Achtung:

Sie haben Javascript deaktiviert!
Sie haben versucht eine Funktion zu nutzen, die nur mit Javascript möglich ist. Um sämtliche Funktionalitäten unserer Internetseite zu nutzen, aktivieren Sie bitte Javascript in Ihrem Browser.

Info-Icon Diese Seite ist nicht in Deutsch verfügbar

Universität
Paderborn
Data Center Building O Bildinformationen anzeigen

Data Center Building O

PC² Publications


Liste im Research Information System öffnen

2023

A computation of D(9) using FPGA Supercomputing

L. Van Hirtum, P. De Causmaecker, J. Goemaere, T. Kenter, H. Riebler, M. Lass, C. Plessl, in: arXiv:2304.03039, 2023

This preprint makes the claim of having computed the $9^{th}$ Dedekind Number. This was done by building an efficient FPGA Accelerator for the core operation of the process, and parallelizing it on the Noctua 2 Supercluster at Paderborn University. The resulting value is 286386577668298411128469151667598498812366. This value can be verified in two steps. We have made the data file containing the 490M results available, each of which can be verified separately on CPU, and the whole file sums to our proposed value.


Computing and Compressing Electron Repulsion Integrals on FPGAs

X. Wu, T. Kenter, R. Schade, T. Kühne, C. Plessl, in: arXiv:2303.13632, 2023

The computation of electron repulsion integrals (ERIs) over Gaussian-type orbitals (GTOs) is a challenging problem in quantum-mechanics-based atomistic simulations. In practical simulations, several trillions of ERIs may have to be computed for every time step. In this work, we investigate FPGAs as accelerators for the ERI computation. We use template parameters, here within the Intel oneAPI tool flow, to create customized designs for 256 different ERI quartet classes, based on their orbitals. To maximize data reuse, all intermediates are buffered in FPGA on-chip memory with customized layout. The pre-calculation of intermediates also helps to overcome data dependencies caused by multi-dimensional recurrence relations. The involved loop structures are partially or even fully unrolled for high throughput of FPGA kernels. Furthermore, a lossy compression algorithm utilizing arbitrary bitwidth integers is integrated in the FPGA kernels. To our best knowledge, this is the first work on ERI computation on FPGAs that supports more than just the single most basic quartet class. Also, the integration of ERI computation and compression it a novelty that is not even covered by CPU or GPU libraries so far. Our evaluation shows that using 16-bit integer for the ERI compression, the fastest FPGA kernels exceed the performance of 10 GERIS ($10 \times 10^9$ ERIs per second) on one Intel Stratix 10 GX 2800 FPGA, with maximum absolute errors around $10^{-7}$ - $10^{-5}$ Hartree. The measured throughput can be accurately explained by a performance model. The FPGA kernels deployed on 2 FPGAs outperform similar computations using the widely used libint reference on a two-socket server with 40 Xeon Gold 6148 CPU cores of the same process technology by factors up to 6.0x and on a new two-socket server with 128 EPYC 7713 CPU cores by up to 1.9x.


Multi-FPGA Designs and Scaling of HPC Challenge Benchmarks via MPI and Circuit-Switched Inter-FPGA Networks

M. Meyer, T. Kenter, C. Plessl, ACM Transactions on Reconfigurable Technology and Systems (2023)

<jats:p>While FPGA accelerator boards and their respective high-level design tools are maturing, there is still a lack of multi-FPGA applications, libraries, and not least, benchmarks and reference implementations towards sustained HPC usage of these devices. As in the early days of GPUs in HPC, for workloads that can reasonably be decoupled into loosely coupled working sets, multi-accelerator support can be achieved by using standard communication interfaces like MPI on the host side. However, for performance and productivity, some applications can profit from a tighter coupling of the accelerators. FPGAs offer unique opportunities here when extending the dataflow characteristics to their communication interfaces.</jats:p> <jats:p>In this work, we extend the HPCC FPGA benchmark suite by multi-FPGA support and three missing benchmarks that particularly characterize or stress inter-device communication: b_eff, PTRANS, and LINPACK. With all benchmarks implemented for current boards with Intel and Xilinx FPGAs, we established a baseline for multi-FPGA performance. Additionally, for the communication-centric benchmarks, we explored the potential of direct FPGA-to-FPGA communication with a circuit-switched inter-FPGA network that is currently only available for one of the boards. The evaluation with parallel execution on up to 26 FPGA boards makes use of one of the largest academic FPGA installations.</jats:p>


2022

Breaking the Exascale Barrier for the Electronic Structure Problem in Ab-Initio Molecular Dynamics

R. Schade, T. Kenter, H. Elgabarty, M. Lass, T. Kühne, C. Plessl, in: arXiv:2205.12182, 2022

The non-orthogonal local submatrix method applied to electronic-structure based molecular dynamics simulations is shown to exceed 1.1 EFLOP/s in FP16/FP32 mixed floating-point arithmetic when using 4,400 NVIDIA A100 GPUs of the Perlmutter system. This is enabled by a modification of the original method that pushes the sustained fraction of the peak performance to about 80%. Example calculations are performed for SARS-CoV-2 spike proteins with up to 83 million atoms.


Bridging HPC Communities through the Julia Programming Language

V. Churavy, W.F. Godoy, C. Bauer, H. Ranocha, M. Schlottke-Lakemper, L. Räss, J. Blaschke, M. Giordano, E. Schnetter, S. Omlin, J.S. Vetter, A. Edelman, 2022

The Julia programming language has evolved into a modern alternative to fill existing gaps in scientific computing and data science applications. Julia leverages a unified and coordinated single-language and ecosystem paradigm and has a proven track record of achieving high performance without sacrificing user productivity. These aspects make Julia a viable alternative to high-performance computing's (HPC's) existing and increasingly costly many-body workflow composition strategy in which traditional HPC languages (e.g., Fortran, C, C++) are used for simulations, and higher-level languages (e.g., Python, R, MATLAB) are used for data analysis and interactive computing. Julia's rapid growth in language capabilities, package ecosystem, and community make it a promising universal language for HPC. This paper presents the views of a multidisciplinary group of researchers from academia, government, and industry that advocate for an HPC software development paradigm that emphasizes developer productivity, workflow portability, and low barriers for entry. We believe that the Julia programming language, its ecosystem, and its community provide modern and powerful capabilities that enable this group's objectives. Crucially, we believe that Julia can provide a feasible and less costly approach to programming scientific applications and workflows that target HPC facilities. In this work, we examine the current practice and role of Julia as a common, end-to-end programming model to address major challenges in scientific reproducibility, data-driven AI/machine learning, co-design and workflows, scalability and performance portability in heterogeneous computing, network communication, data management, and community education. As a result, the diversification of current investments to fulfill the needs of the upcoming decade is crucial as more supercomputing centers prepare for the exascale era.



CP2K on the road to exascale

T. Kühne, C. Plessl, R. Schade, O. Schütt, in: arXiv:2205.14741, 2022

The CP2K program package, which can be considered as the swiss army knife of atomistic simulations, is presented with a special emphasis on ab-initio molecular dynamics using the second-generation Car-Parrinello method. After outlining current and near-term development efforts with regards to massively parallel low-scaling post-Hartree-Fock and eigenvalue solvers, novel approaches on how we plan to take full advantage of future low-precision hardware architectures are introduced. Our focus here is on combining our submatrix method with the approximate computing paradigm to address the immanent exascale era.


Dataset for random uniform distributions of 2D circles and 3D spheres.

M. Wojciechowski, Data Brief (2022), 43, pp. 108318


Early time behavior of spatial and momentum anisotropies in kinetic theory across different Knudsen numbers

N. Borghini, M. Borrell, H. Roch, in: arXiv:2201.13294, 2022

We investigate the early time development of the anisotropic transverse flow and spatial eccentricities of a fireball with various particle-based transport approaches using a fixed initial condition. In numerical simulations ranging from the quasi-collisionless case to the hydrodynamic regime, we find that the onset of $v_n$ and of related measures of anisotropic flow can be described with a simple power-law ansatz, with an exponent that depends on the amount of rescatterings in the system. In the few-rescatterings regime we perform semi-analytical calculations, based on a systematic expansion in powers of time and the cross section, which can reproduce the numerical findings.


Even anisotropic-flow harmonics are from Venus, odd ones are from Mars

B. Bachmann, N. Borghini, N. Feld, H. Roch, in: arXiv:2203.13306, 2022

We test the ability of the "escape mechanism" to create the anisotropic flow observed in high-energy nuclear collisions. We compare the flow harmonics $v_n$ in the few-rescatterings regime from two types of transport simulations, with $2\to 2$ and $2\to 0$ collision kernels respectively, and from analytical calculations neglecting the gain term of the Boltzmann equation. We find that the even flow harmonics are similar in the three approaches, while the odd harmonics differ significantly.


In-depth FPGA Accelerator Performance Evaluation with Single Node Benchmarks from the HPC Challenge Benchmark Suite for Intel and Xilinx FPGAs using OpenCL

M. Meyer, T. Kenter, C. Plessl, Journal of Parallel and Distributed Computing (2022)

DOI


Linearly shifting ferromagnetic resonance response of La0.7Sr0.3MnO3 thin film for body temperature sensors

W. Hou, Y. Yao, Y. Li, B. Peng, K. Shi, Z. Zhou, J. Pan, M. Liu, J. Hu, Frontiers of materials science (2022), 16(1)


Nonlinear down-conversion in a single quantum dot

B. Jonas, D.F. Heinze, E. Schöll, P. Kallert, T. Langer, S. Krehs, A. Widhalm, K. Jöns, D. Reuter, S. Schumacher, A. Zrenner, Nature Communications (2022), 13(1), 1387

<jats:title>Abstract</jats:title><jats:p>Tailored nanoscale quantum light sources, matching the specific needs of use cases, are crucial building blocks for photonic quantum technologies. Several different approaches to realize solid-state quantum emitters with high performance have been pursued and different concepts for energy tuning have been established. However, the properties of the emitted photons are always defined by the individual quantum emitter and can therefore not be controlled with full flexibility. Here we introduce an all-optical nonlinear method to tailor and control the single photon emission. We demonstrate a laser-controlled down-conversion process from an excited state of a semiconductor quantum three-level system. Based on this concept, we realize energy tuning and polarization control of the single photon emission with a control-laser field. Our results mark an important step towards tailored single photon emission from a photonic quantum system based on quantum optical principles.</jats:p>


Parallel quantum chemistry on noisy intermediate-scale quantum computers

R. Schade, C. Bauer, K. Tamoev, L. Mazur, C. Plessl, T. Kühne, Phys. Rev. Research (2022), 4, pp. 033160

A parallel hybrid quantum-classical algorithm for the solution of the quantum-chemical ground-state energy problem on gate-based quantum computers is presented. This approach is based on the reduced density-matrix functional theory (RDMFT) formulation of the electronic structure problem. For that purpose, the density-matrix functional of the full system is decomposed into an indirectly coupled sum of density-matrix functionals for all its subsystems using the adaptive cluster approximation to RDMFT. The approximations involved in the decomposition and the adaptive cluster approximation itself can be systematically converged to the exact result. The solutions for the density-matrix functionals of the effective subsystems involves a constrained minimization over many-particle states that are approximated by parametrized trial states on the quantum computer similarly to the variational quantum eigensolver. The independence of the density-matrix functionals of the effective subsystems introduces a new level of parallelization and allows for the computational treatment of much larger molecules on a quantum computer with a given qubit count. In addition, for the proposed algorithm techniques are presented to reduce the qubit count, the number of quantum programs, as well as its depth. The evaluation of a density-matrix functional as the essential part of our approach is demonstrated for Hubbard-like systems on IBM quantum computers based on superconducting transmon qubits.


Roadmap on Electronic Structure Codes in the Exascale Era

V. Gavini, S. Baroni, V. Blum, D.R. Bowler, A. Buccheri, J.R. Chelikowsky, S. Das, W. Dawson, P. Delugas, M. Dogan, C. Draxl, G. Galli, L. Genovese, P. Giannozzi, M. Giantomassi, X. Gonze, M. Govoni, A. Gulans, F. Gygi, J.M. Herbert, S. Kokott, T. Kühne, K. Liou, T. Miyazaki, P. Motamarri, A. Nakata, J.E. Pask, C. Plessl, L.E. Ratcliff, R.M. Richard, M. Rossi, R. Schade, M. Scheffler, O. Schütt, P. Suryanarayana, M. Torrent, L. Truflandier, T.L. Windus, Q. Xu, V.W.-. Yu, D. Perez, in: arXiv:2209.12747, 2022

Electronic structure calculations have been instrumental in providing many important insights into a range of physical and chemical properties of various molecular and solid-state systems. Their importance to various fields, including materials science, chemical sciences, computational chemistry and device physics, is underscored by the large fraction of available public supercomputing resources devoted to these calculations. As we enter the exascale era, exciting new opportunities to increase simulation numbers, sizes, and accuracies present themselves. In order to realize these promises, the community of electronic structure software developers will however first have to tackle a number of challenges pertaining to the efficient use of new architectures that will rely heavily on massive parallelism and hardware accelerators. This roadmap provides a broad overview of the state-of-the-art in electronic structure calculations and of the various new directions being pursued by the community. It covers 14 electronic structure codes, presenting their current status, their development priorities over the next five years, and their plans towards tackling the challenges and leveraging the opportunities presented by the advent of exascale computing.


2021

Diffusion of the carbon dioxide–ethanol mixture in the extended critical region

R.S. Chatwell, G. Guevara-Carrion, Y. Gaponenko, V. Shevtsova, J. Vrabec, Physical Chemistry Chemical Physics (2021), 23(4), pp. 3106-3115

<p>The effect of traces of ethanol in supercritical carbon dioxide on the mixture's thermodynamic properties is studied by molecular simulations and Taylor dispersion measurements.</p>


Evaluating the Design Space for Offloading 3D FFT Calculations to an FPGA for High-Performance Computing

A. Ramaswami, T. Kenter, T. Kühne, C. Plessl, in: Applied Reconfigurable Computing. Architectures, Tools, and Applications, Springer International Publishing, 2021

DOI


Generating Physically Sound Training Data for Image Recognition of Additively Manufactured Parts

T. Nickchen, S. Heindorf, G. Engels, in: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, 2021, pp. 1994-2002


High-Performance Spectral Element Methods on Field-Programmable Gate Arrays : Implementation, Evaluation, and Future Projection

M. Karp, A. Podobas, N. Jansson, T. Kenter, C. Plessl, P. Schlatter, S. Markidis, in: 2021 IEEE International Parallel and Distributed Processing Symposium (IPDPS), IEEE, 2021

DOI


HighPerMeshes – A Domain-Specific Language for Numerical Algorithms on Unstructured Grids

S. Alhaddad, J. Förstner, S. Groth, D. Grünewald, Y. Grynko, F. Hannig, T. Kenter, F. Pfreundt, C. Plessl, M. Schotte, T. Steinke, J. Teich, M. Weiser, F. Wende, in: Euro-Par 2020: Parallel Processing Workshops, 2021

Solving partial differential equations on unstructured grids is a cornerstone of engineering and scientific computing. Nowadays, heterogeneous parallel platforms with CPUs, GPUs, and FPGAs enable energy-efficient and computationally demanding simulations. We developed the HighPerMeshes C++-embedded Domain-Specific Language (DSL) for bridging the abstraction gap between the mathematical and algorithmic formulation of mesh-based algorithms for PDE problems on the one hand and an increasing number of heterogeneous platforms with their different parallel programming and runtime models on the other hand. Thus, the HighPerMeshes DSL aims at higher productivity in the code development process for multiple target platforms. We introduce the concepts as well as the basic structure of the HighPerMeshes DSL, and demonstrate its usage with three examples, a Poisson and monodomain problem, respectively, solved by the continuous finite element method, and the discontinuous Galerkin method for Maxwell’s equation. The mapping of the abstract algorithmic description onto parallel hardware, including distributed memory compute clusters, is presented. Finally, the achievable performance and scalability are demonstrated for a typical example problem on a multi-core CPU cluster.


How irreversible are steady-state trajectories of a trapped active particle?

L. Dabelow, S. Bo, R. Eichhorn, Journal of Statistical Mechanics: Theory and Experiment (2021), 2021(3), 033216

<jats:title>Abstract</jats:title> <jats:p>The defining feature of active particles is that they constantly propel themselves by locally converting chemical energy into directed motion. This active self-propulsion prevents them from equilibrating with their thermal environment (e.g. an aqueous solution), thus keeping them permanently out of equilibrium. Nevertheless, the spatial dynamics of active particles might share certain equilibrium features, in particular in the steady state. We here focus on the time-reversal symmetry of individual spatial trajectories as a distinct equilibrium characteristic. We investigate to what extent the steady-state trajectories of a trapped active particle obey or break this time-reversal symmetry. Within the framework of active Ornstein–Uhlenbeck particles we find that the steady-state trajectories in a harmonic potential fulfill path-wise time-reversal symmetry exactly, while this symmetry is typically broken in anharmonic potentials.</jats:p>


OpenCL-based FPGA Accelerator for Semi-Global Approximate String Matching Using Diagonal Bit-Vectors

D. Castells-Rufas, S. Marco-Sola, Q. Aguado-Puig, A. Espinosa-Morales, J.C. Moure, L. Alvarez, M. Moreto, in: 2021 31st International Conference on Field-Programmable Logic and Applications (FPL), IEEE, 2021

An FPGA accelerator for the computation of the semi-global Levenshtein distance between a pattern and a reference text is presented. The accelerator provides an important benefit to reduce the execution time of read-mappers used in short-read genomic sequencing. Previous attempts to solve the same problem in FPGA use the Myers algorithm following a column approach to compute the dynamic programming table. We use an approach based on diagonals that allows for some resource savings while maintaining a very high throughput of 1 alignment per clock cycle. The design is implemented in OpenCL and tested on two FPGA accelerators. The maximum performance obtained is 91.5 MPairs/s for 100 × 120 sequences and 47 MPairs/s for 300 × 360 sequences, the highest ever reported for this problem.


Optimization of optical waveguide antennas for directive emission of light

H. Farheen, T. Leuteritz, S. Linden, V. Myroshnychenko, J. Förstner, in: arXiv:2106.02468, 2021

Optical travelling wave antennas offer unique opportunities to control and selectively guide light into a specific direction which renders them as excellent candidates for optical communication and sensing. These applications require state of the art engineering to reach optimized functionalities such as high directivity and radiation efficiency, low side lobe level, broadband and tunable capabilities, and compact design. In this work we report on the numerical optimization of the directivity of optical travelling wave antennas made from low-loss dielectric materials using full-wave numerical simulations in conjunction with a particle swarm optimization algorithm. The antennas are composed of a reflector and a director deposited on a glass substrate and an emitter placed in the feed gap between them serves as an internal source of excitation. In particular, we analysed antennas with rectangular- and horn-shaped directors made of either Hafnium dioxide or Silicon. The optimized antennas produce highly directional emission due to the presence of two dominant guided TE modes in the director in addition to leaky modes. These guided modes dominate the far-field emission pattern and govern the direction of the main lobe emission which predominately originates from the end facet of the director. Our work also provides a comprehensive analysis of the modes, radiation patterns, parametric influences, and bandwidths of the antennas that highlights their robust nature.


Steady states of $Λ$-type three-level systems excited by quantum light in lossy cavities

H. Rose, O.V. Tikhonova, T. Meier, P.. Sharapova, in: arXiv:2109.00842, 2021

The interaction between quantum light and matter is being intensively studied for systems that are enclosed in high-$Q$ cavities which strongly enhance the light-matter coupling. However, for many applications, cavities with lower $Q$-factors are preferred due to the increased spectral width of the cavity mode. Here, we investigate the interaction between quantum light and matter represented by a $\Lambda$-type three-level system in lossy cavities, assuming that cavity losses are the dominant loss mechanism. We demonstrate that cavity losses lead to non-trivial steady states of the electronic occupations that can be controlled by the loss rate and the initial statistics of the quantum fields. The mechanism of formation of such steady states can be understood on the basis of the equations of motion. Analytical expressions for steady states and their numerical simulations are presented and discussed.


The HighPerMeshes framework for numerical algorithms on unstructured grids

S. Alhaddad, J. Förstner, S. Groth, D. Grünewald, Y. Grynko, F. Hannig, T. Kenter, F. Pfreundt, C. Plessl, M. Schotte, T. Steinke, J. Teich, M. Weiser, F. Wende, Concurrency and Computation: Practice and Experience (2021), pp. e6616


The Strong Scaling Advantage of FPGAs in HPC for N-body Simulations

J. Menzel, C. Plessl, T. Kenter, ACM Transactions on Reconfigurable Technology and Systems (2021), 15(1), pp. 1-30

N-body methods are one of the essential algorithmic building blocks of high-performance and parallel computing. Previous research has shown promising performance for implementing n-body simulations with pairwise force calculations on FPGAs. However, to avoid challenges with accumulation and memory access patterns, the presented designs calculate each pair of forces twice, along with both force sums of the involved particles. Also, they require large problem instances with hundreds of thousands of particles to reach their respective peak performance, limiting the applicability for strong scaling scenarios. This work addresses both issues by presenting a novel FPGA design that uses each calculated force twice and overlaps data transfers and computations in a way that allows to reach peak performance even for small problem instances, outperforming previous single precision results even in double precision, and scaling linearly over multiple interconnected FPGAs. For a comparison across architectures, we provide an equally optimized CPU reference, which for large problems actually achieves higher peak performance per device, however, given the strong scaling advantages of the FPGA design, in parallel setups with few thousand particles per device, the FPGA platform achieves highest performance and power efficiency.


Towards Electronic Structure-Based Ab-Initio Molecular Dynamics Simulations with Hundreds of Millions of Atoms

R. Schade, T. Kenter, H. Elgabarty, M. Lass, O. Schütt, A. Lazzaro, H. Pabst, S. Mohr, J. Hutter, T.D. Kühne, C. Plessl, in: arXiv:2104.08245, 2021

We push the boundaries of electronic structure-based \textit{ab-initio} molecular dynamics (AIMD) beyond 100 million atoms. This scale is otherwise barely reachable with classical force-field methods or novel neural network and machine learning potentials. We achieve this breakthrough by combining innovations in linear-scaling AIMD, efficient and approximate sparse linear algebra, low and mixed-precision floating-point computation on GPUs, and a compensation scheme for the errors introduced by numerical approximations. The core of our work is the non-orthogonalized local submatrix method (NOLSM), which scales very favorably to massively parallel computing systems and translates large sparse matrix operations into highly parallel, dense matrix operations that are ideally suited to hardware accelerators. We demonstrate that the NOLSM method, which is at the center point of each AIMD step, is able to achieve a sustained performance of 324 PFLOP/s in mixed FP16/FP32 precision corresponding to an efficiency of 67.7% when running on 1536 NVIDIA A100 GPUs.


Towards Performance Characterization of FPGAs in Context of HPC using OpenCL Benchmarks

M. Meyer, in: Proceedings of the 11th International Symposium on Highly Efficient Accelerators and Reconfigurable Technologies, 2021

DOI


2020

<i>In silico</i> investigation of Cu(In,Ga)Se<sub>2</sub>-based solar cells

H. Mirhosseini, R. Kormath Madam Raghupathy, S.K. Sahoo, H. Wiebeler, M. Chugh, T.D. Kühne, Physical Chemistry Chemical Physics (2020), 22(46), pp. 26682-26701

<p>State-of-the-art methods in materials science such as artificial intelligence and data-driven techniques advance the investigation of photovoltaic materials.</p>


A Runtime System for Finite Element Methods in a Partitioned Global Address Space

S. Groth, D. Grünewald, J. Teich, F. Hannig, in: Proceedings of the 17th ACM International Conference on Computing Frontiers (CF '2020), ACM, 2020

DOI


A Submatrix-Based Method for Approximate Matrix Function Evaluation in the Quantum Chemistry Code CP2K

M. Lass, R. Schade, T. Kühne, C. Plessl, in: Proc. International Conference for High Performance Computing, Networking, Storage and Analysis (SC), IEEE Computer Society, 2020, pp. 1127-1140

Electronic structure calculations based on density-functional theory (DFT) represent a significant part of today's HPC workloads and pose high demands on high-performance computing resources. To perform these quantum-mechanical DFT calculations on complex large-scale systems, so-called linear scaling methods instead of conventional cubic scaling methods are required. In this work, we take up the idea of the submatrix method and apply it to the DFT computations in the software package CP2K. For that purpose, we transform the underlying numeric operations on distributed, large, sparse matrices into computations on local, much smaller and nearly dense matrices. This allows us to exploit the full floating-point performance of modern CPUs and to make use of dedicated accelerator hardware, where performance has been limited by memory bandwidth before. We demonstrate both functionality and performance of our implementation and show how it can be accelerated with GPUs and FPGAs.


Accurate Sampling with Noisy Forces from Approximate Computing

V. Rengaraj, M. Lass, C. Plessl, T. Kühne, Computation (2020), 8(2), 39

In scientific computing, the acceleration of atomistic computer simulations by means of custom hardware is finding ever-growing application. A major limitation, however, is that the high efficiency in terms of performance and low power consumption entails the massive usage of low precision computing units. Here, based on the approximate computing paradigm, we present an algorithmic method to compensate for numerical inaccuracies due to low accuracy arithmetic operations rigorously, yet still obtaining exact expectation values using a properly modified Langevin-type equation.


CP2K: An electronic structure and molecular dynamics software package - Quickstep: Efficient and accurate electronic structure calculations

T. Kühne, M. Iannuzzi, M.D. Ben, V.V. Rybkin, P. Seewald, F. Stein, T. Laino, R.Z. Khaliullin, O. Schütt, F. Schiffmann, D. Golze, J. Wilhelm, S. Chulkov, M.H.B. Mohammad Hossein Bani-Hashemian, V. Weber, U. Borstnik, M. Taillefumier, A.S. Jakobovits, A. Lazzaro, H. Pabst, T. Müller, R. Schade, M. Guidon, S. Andermatt, N. Holmberg, G.K. Schenter, A. Hehn, A. Bussy, F. Belleflamme, G. Tabacchi, A. Glöß, M. Lass, I. Bethune, C.J. Mundy, C. Plessl, M. Watkins, J. VandeVondele, M. Krack, J. Hutter, The Journal of Chemical Physics (2020), 152(19), 194103

CP2K is an open source electronic structure and molecular dynamics software package to perform atomistic simulations of solid-state, liquid, molecular, and biological systems. It is especially aimed at massively parallel and linear-scaling electronic structure methods and state-of-theart ab initio molecular dynamics simulations. Excellent performance for electronic structure calculations is achieved using novel algorithms implemented for modern high-performance computing systems. This review revisits the main capabilities of CP2K to perform efficient and accurate electronic structure simulations. The emphasis is put on density functional theory and multiple post–Hartree–Fock methods using the Gaussian and plane wave approach and its augmented all-electron extension.


Evaluating FPGA Accelerator Performance with a Parameterized OpenCL Adaptation of Selected Benchmarks of the HPCChallenge Benchmark Suite

M. Meyer, T. Kenter, C. Plessl, in: 2020 IEEE/ACM International Workshop on Heterogeneous High-performance Reconfigurable Computing (H2RC), 2020

FPGAs have found increasing adoption in data center applications since a new generation of high-level tools have become available which noticeably reduce development time for FPGA accelerators and still provide high-quality results. There is, however, no high-level benchmark suite available, which specifically enables a comparison of FPGA architectures, programming tools, and libraries for HPC applications. To fill this gap, we have developed an OpenCL-based open-source implementation of the HPCC benchmark suite for Xilinx and Intel FPGAs. This benchmark can serve to analyze the current capabilities of FPGA devices, cards, and development tool flows, track progress over time, and point out specific difficulties for FPGA acceleration in the HPC domain. Additionally, the benchmark documents proven performance optimization patterns. We will continue optimizing and porting the benchmark for new generations of FPGAs and design tools and encourage active participation to create a valuable tool for the community. To fill this gap, we have developed an OpenCL-based open-source implementation of the HPCC benchmark suite for Xilinx and Intel FPGAs. This benchmark can serve to analyze the current capabilities of FPGA devices, cards, and development tool flows, track progress over time, and point out specific difficulties for FPGA acceleration in the HPC domain. Additionally, the benchmark documents proven performance optimization patterns. We will continue optimizing and porting the benchmark for new generations of FPGAs and design tools and encourage active participation to create a valuable tool for the community.


Multi-Armed Bandits with Censored Consumption of Resources

V. Bengs, E. Hüllermeier, in: arXiv:2011.00813, 2020

We consider a resource-aware variant of the classical multi-armed bandit problem: In each round, the learner selects an arm and determines a resource limit. It then observes a corresponding (random) reward, provided the (random) amount of consumed resources remains below the limit. Otherwise, the observation is censored, i.e., no reward is obtained. For this problem setting, we introduce a measure of regret, which incorporates the actual amount of allocated resources of each learning round as well as the optimality of realizable rewards. Thus, to minimize regret, the learner needs to set a resource limit and choose an arm in such a way that the chance to realize a high reward within the predefined resource limit is high, while the resource limit itself should be kept as low as possible. We derive the theoretical lower bound on the cumulative regret and propose a learning algorithm having a regret upper bound that matches the lower bound. In a simulation study, we show that our learning algorithm outperforms straightforward extensions of standard multi-armed bandit algorithms.



2019

A General Algorithm to Calculate the Inverse Principal p-th Root of Symmetric Positive Definite Matrices

D. Richters, M. Lass, A. Walther, C. Plessl, T. Kühne, Communications in Computational Physics (2019), 25(2), pp. 564-585

We address the general mathematical problem of computing the inverse p-th root of a given matrix in an efficient way. A new method to construct iteration functions that allow calculating arbitrary p-th roots and their inverses of symmetric positive definite matrices is presented. We show that the order of convergence is at least quadratic and that adaptively adjusting a parameter q always leads to an even faster convergence. In this way, a better performance than with previously known iteration schemes is achieved. The efficiency of the iterative functions is demonstrated for various matrices with different densities, condition numbers and spectral radii.



FPGAs im Rechenzentrum

M. Platzner, C. Plessl, Informatik Spektrum (2019)


OpenCL Implementation of Cannon's Matrix Multiplication Algorithm on Intel Stratix 10 FPGAs

P. Gorlani, T. Kenter, C. Plessl, in: Proceedings of the International Conference on Field-Programmable Technology (FPT), IEEE, 2019

Stratix 10 FPGA cards have a good potential for the acceleration of HPC workloads since the Stratix 10 product line introduces devices with a large number of DSP and memory blocks. The high level synthesis of OpenCL codes can play a fundamental role for FPGAs in HPC, because it allows to implement different designs with lower development effort compared to hand optimized HDL. However, Stratix 10 cards are still hard to fully exploit using the Intel FPGA SDK for OpenCL. The implementation of designs with thousands of concurrent arithmetic operations often suffers from place and route problems that limit the maximum frequency or entirely prevent a successful synthesis. In order to overcome these issues for the implementation of the matrix multiplication, we formulate Cannon's matrix multiplication algorithm with regard to its efficient synthesis within the FPGA logic. We obtain a two-level block algorithm, where the lower level sub-matrices are multiplied using our Cannon's algorithm implementation. Following this design approach with multiple compute units, we are able to get maximum frequencies close to and above 300 MHz with high utilization of DSP and memory blocks. This allows for performance results above 1 TeraFLOPS.


Transparent Acceleration for Heterogeneous Platforms with Compilation to OpenCL

H. Riebler, G.F. Vaz, T. Kenter, C. Plessl, ACM Trans. Archit. Code Optim. (TACO) (2019), 16(2), pp. 14:1–14:26

DOI


2018

A Data Structure for Planning Based Workload Management of Heterogeneous HPC Systems

A. Keller, in: Proc. Workshop on Job Scheduling Strategies for Parallel Processing (JSSPP), Springer, 2018, pp. 132-151

This paper describes a data structure and a heuristic to plan and map arbitrary resources in complex combinations while applying time dependent constraints. The approach is used in the planning based workload manager OpenCCS at the Paderborn Center for Parallel Computing (PC\(^2\)) to operate heterogeneous clusters with up to 10000 cores. We also show performance results derived from four years of operation.


A Massively Parallel Algorithm for the Approximate Calculation of Inverse p-th Roots of Large Sparse Matrices

M. Lass, S. Mohr, H. Wiebeler, T. Kühne, C. Plessl, in: Proc. Platform for Advanced Scientific Computing (PASC) Conference, ACM, 2018

We present the submatrix method, a highly parallelizable method for the approximate calculation of inverse p-th roots of large sparse symmetric matrices which are required in different scientific applications. Following the idea of Approximate Computing, we allow imprecision in the final result in order to utilize the sparsity of the input matrix and to allow massively parallel execution. For an n x n matrix, the proposed algorithm allows to distribute the calculations over n nodes with only little communication overhead. The result matrix exhibits the same sparsity pattern as the input matrix, allowing for efficient reuse of allocated data structures. We evaluate the algorithm with respect to the error that it introduces into calculated results, as well as its performance and scalability. We demonstrate that the error is relatively limited for well-conditioned matrices and that results are still valuable for error-resilient applications like preconditioning even for ill-conditioned matrices. We discuss the execution time and scaling of the algorithm on a theoretical level and present a distributed implementation of the algorithm using MPI and OpenMP. We demonstrate the scalability of this implementation by running it on a high-performance compute cluster comprised of 1024 CPU cores, showing a speedup of 665x compared to single-threaded execution.


Automated Code Acceleration Targeting Heterogeneous OpenCL Devices

H. Riebler, G.F. Vaz, T. Kenter, C. Plessl, in: Proc. ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming (PPoPP), ACM, 2018

DOI


OpenCL-based FPGA Design to Accelerate the Nodal Discontinuous Galerkin Method for Unstructured Meshes

T. Kenter, G. Mahale, S. Alhaddad, Y. Grynko, C. Schmitt, A. Afzal, F. Hannig, J. Förstner, C. Plessl, in: Proc. Int. Symp. on Field-Programmable Custom Computing Machines (FCCM), IEEE, 2018

The exploration of FPGAs as accelerators for scientific simulations has so far mostly been focused on small kernels of methods working on regular data structures, for example in the form of stencil computations for finite difference methods. In computational sciences, often more advanced methods are employed that promise better stability, convergence, locality and scaling. Unstructured meshes are shown to be more effective and more accurate, compared to regular grids, in representing computation domains of various shapes. Using unstructured meshes, the discontinuous Galerkin method preserves the ability to perform explicit local update operations for simulations in the time domain. In this work, we investigate FPGAs as target platform for an implementation of the nodal discontinuous Galerkin method to find time-domain solutions of Maxwell's equations in an unstructured mesh. When maximizing data reuse and fitting constant coefficients into suitably partitioned on-chip memory, high computational intensity allows us to implement and feed wide data paths with hundreds of floating point operators. By decoupling off-chip memory accesses from the computations, high memory bandwidth can be sustained, even for the irregular access pattern required by parts of the application. Using the Intel/Altera OpenCL SDK for FPGAs, we present different implementation variants for different polynomial orders of the method. In different phases of the algorithm, either computational or bandwidth limits of the Arria 10 platform are almost reached, thus outperforming a highly multithreaded CPU implementation by around 2x.


Sprint diagnostic with GPS and inertial sensor fusion

J.C. Mertens, A. Boschmann, M. Schmidt, C. Plessl, Sports Engineering (2018), 21(4), pp. 441-451

DOI


Theory of optically controlled anisotropic polariton transport in semiconductor double microcavities

S.M.H. Luk, P. Lewandowski, N.H. Kwong, E. Baudin, O. Lafont, J. Tignon, P.T. Leung, C.K.P. Chan, M. Babilon, S. Schumacher, R. Binder, Journal of the Optical Society of America B (2018), 35(1), 146

DOI


Using Approximate Computing for the Calculation of Inverse Matrix p-th Roots

M. Lass, T. Kühne, C. Plessl, Embedded Systems Letters (2018), 10(2), pp. 33-36

Approximate computing has shown to provide new ways to improve performance and power consumption of error-resilient applications. While many of these applications can be found in image processing, data classification or machine learning, we demonstrate its suitability to a problem from scientific computing. Utilizing the self-correcting behavior of iterative algorithms, we show that approximate computing can be applied to the calculation of inverse matrix p-th roots which are required in many applications in scientific computing. Results show great opportunities to reduce the computational effort and bandwidth required for the execution of the discussed algorithm, especially when targeting special accelerator hardware.


2017

Efficient Branch and Bound on FPGAs Using Work Stealing and Instance-Specific Designs

H. Riebler, M. Lass, R. Mittendorf, T. Löcke, C. Plessl, ACM Transactions on Reconfigurable Technology and Systems (TRETS) (2017), 10(3), pp. 24:1-24:23

Branch and bound (B&B) algorithms structure the search space as a tree and eliminate infeasible solutions early by pruning subtrees that cannot lead to a valid or optimal solution. Custom hardware designs significantly accelerate the execution of these algorithms. In this article, we demonstrate a high-performance B&B implementation on FPGAs. First, we identify general elements of B&B algorithms and describe their implementation as a finite state machine. Then, we introduce workers that autonomously cooperate using work stealing to allow parallel execution and full utilization of the target FPGA. Finally, we explore advantages of instance-specific designs that target a specific problem instance to improve performance. We evaluate our concepts by applying them to a branch and bound problem, the reconstruction of corrupted AES keys obtained from cold-boot attacks. The evaluation shows that our work stealing approach is scalable with the available resources and provides speedups proportional to the number of workers. Instance-specific designs allow us to achieve an overall speedup of 47 × compared to the fastest implementation of AES key reconstruction so far. Finally, we demonstrate how instance-specific designs can be generated just-in-time such that the provided speedups outweigh the additional time required for design synthesis.


Flexible FPGA design for FDTD using OpenCL

T. Kenter, J. Förstner, C. Plessl, in: Proc. Int. Conf. on Field Programmable Logic and Applications (FPL), IEEE, 2017

Compared to classical HDL designs, generating FPGA with high-level synthesis from an OpenCL specification promises easier exploration of different design alternatives and, through ready-to-use infrastructure and common abstractions for host and memory interfaces, easier portability between different FPGA families. In this work, we evaluate the extent of this promise. To this end, we present a parameterized FDTD implementation for photonic microcavity simulations. Our design can trade-off different forms of parallelism and works for two independent OpenCL-based FPGA design flows. Hence, we can target FPGAs from different vendors and different FPGA families. We describe how we used pre-processor macros to achieve this flexibility and to work around different shortcomings of the current tools. Choosing the right design configurations, we are able to present two extremely competitive solutions for very different FPGA targets, reaching up to 172 GFLOPS sustained performance. With the portability and flexibility demonstrated, code developers not only avoid vendor lock-in, but can even make best use of real trade-offs between different architectures.


High-Throughput and Low-Latency Network Communication with NetIO

J. Schumacher, C. Plessl, W. Vandelli, Journal of Physics: Conference Series (2017), 898, 082003

DOI


2016

Confidentiality and Authenticity for Distributed Version Control Systems - A Mercurial Extension

M. Lass, D. Leibenger, C. Sorge, in: Proc. 41st Conference on Local Computer Networks (LCN), IEEE, 2016

Version Control Systems (VCS) are a valuable tool for software development and document management. Both client/server and distributed (Peer-to-Peer) models exist, with the latter (e.g., Git and Mercurial) becoming increasingly popular. Their distributed nature introduces complications, especially concerning security: it is hard to control the dissemination of contents stored in distributed VCS as they rely on replication of complete repositories to any involved user. We overcome this issue by designing and implementing a concept for cryptography-enforced access control which is transparent to the user. Use of field-tested schemes (end-to-end encryption, digital signatures) allows for strong security, while adoption of convergent encryption and content-defined chunking retains storage efficiency. The concept is seamlessly integrated into Mercurial---respecting its distributed storage concept---to ensure practical usability and compatibility to existing deployments.


Microdisk Cavity FDTD Simulation on FPGA using OpenCL

T. Kenter, C. Plessl, in: Proc. Workshop on Heterogeneous High-performance Reconfigurable Computing (H2RC), 2016


Modeling and simulation of metallic, particle-damped spheres for lightweight materials

T. Steinle, 2016

In vielen technischen Anwendungen spielt heute der Leichtbau eine große Rolle, denn durch Gewichtseinsparungen lässt sich auch Energie einsparen. Allerdings birgt der Leichtbau die Gefahr einer erhöhten Störanfälligkeit gegenüber Vibrationen, die durch die Operation von Maschinen entstehen können. Das Fraunhofer Institut für Fertigungstechnik und Angewandte Materialforschung in Dresden beschäftigt sich mit den Möglichkeiten einer Schwingungsdämpfung durch Verbundwerkstoffe. Dabei wird in die Leichtbaustruktur eine Vielzahl von Hohlkugeln eingebracht, die mit Keramikpartikeln gefüllt sind. Diese Fragestellung bildet die technische Motivation für diese Arbeit. Ziel ist, ein Experiment zur Bestimmung des Restitutionskoeffizienten numerisch nachzubilden. Die Simulation basiert auf einer Diskreten Elemente Methode um die Trajektorien der einzelnen Partikel und der Kugel berechnen zu können. Basierend auf einem Potentialansatz für die Interaktionsberechnung in der Molekulardynamik kann das Reibungsverhalten vielfältig angepasst werden. Das Simulationsvolumen wird durch reflektierende Randbedingungen abgeschlossen und umfasst die Kugelhülle. Dazu kam eine hochflexible Speicherstruktur zum Einsatz, um die heterogene Verteilung der Partikel im Raum mit einer effizienten Linked Cell Methode abbilden zu können. Dadurch wird eine in der Partikelzahl lineare Komplexität erreicht. Umfangreiche numerische Experimente zeigen den großen Effekt der Partikelfüllung auf das Dämpfungsverhalten.


Multiobjective Optimal Control Methods for the Development of an Intelligent Cruise Control

M. Dellnitz, J. Eckstein, K. Flaßkamp, P. Friedel, C. Horenkamp, U. Köhler, S. Ober-Blöbaum, S. Peitz, S. Tiemeyer, in: Progress in Industrial Mathematics at ECMI, Springer International Publishing, 2016, pp. 633-641

DOI


Opportunities for deferring application partitioning and accelerator synthesis to runtime (extended abstract)

T. Kenter, G.F. Vaz, H. Riebler, C. Plessl, in: Workshop on Reconfigurable Computing (WRC), 2016


Performance-centric scheduling with task migration for a heterogeneous compute node in the data center

A. Lösch, T. Beisel, T. Kenter, C. Plessl, M. Platzner, in: Proceedings of the 2016 Design, Automation & Test in Europe Conference & Exhibition (DATE), EDA Consortium / IEEE, 2016, pp. 912-917

The use of heterogeneous computing resources, such as Graphic Processing Units or other specialized coprocessors, has become widespread in recent years because of their per- formance and energy efficiency advantages. Approaches for managing and scheduling tasks to heterogeneous resources are still subject to research. Although queuing systems have recently been extended to support accelerator resources, a general solution that manages heterogeneous resources at the operating system- level to exploit a global view of the system state is still missing.In this paper we present a user space scheduler that enables task scheduling and migration on heterogeneous processing resources in Linux. Using run queues for available resources we perform scheduling decisions based on the system state and on task characterization from earlier measurements. With a pro- gramming pattern that supports the integration of checkpoints into applications, we preempt tasks and migrate them between three very different compute resources. Considering static and dynamic workload scenarios, we show that this approach can gain up to 17% performance, on average 7%, by effectively avoiding idle resources. We demonstrate that a work-conserving strategy without migration is no suitable alternative.


Potential and Methods for Embedding Dynamic Offloading Decisions into Application Code

G.F. Vaz, H. Riebler, T. Kenter, C. Plessl, Computers and Electrical Engineering (2016), 55, pp. 91-111

A broad spectrum of applications can be accelerated by offloading computation intensive parts to reconfigurable hardware. However, to achieve speedups, the number of loop it- erations (trip count) needs to be sufficiently large to amortize offloading overheads. Trip counts are frequently not known at compile time, but only at runtime just before entering a loop. Therefore, we propose to generate code for both the CPU and the coprocessor, and defer the offloading decision to the application runtime. We demonstrate how a toolflow, based on the LLVM compiler framework, can automatically embed dynamic offloading de- cisions into the application code. We perform in-depth static and dynamic analysis of pop- ular benchmarks, which confirm the general potential of such an approach. We also pro- pose to optimize the offloading process by decoupling the runtime decision from the loop execution (decision slack). The feasibility of our approach is demonstrated by a toolflow that automatically identifies suitable data-parallel loops and generates code for the FPGA coprocessor of a Convey HC-1. We evaluate the integrated toolflow with representative loops executed for different input data sizes.



ReconOS

A. Agne, M. Platzner, C. Plessl, M. Happe, E. Lübbers, in: FPGAs for Software Programmers, Springer International Publishing, 2016, pp. 227-244

In this chapter, we present an introduction to the ReconOS operating system for reconfigurable computing. ReconOS offers a unified multi-threaded programming model and operating system services for threads executing in software and threads mapped to reconfigurable hardware. By supporting standard POSIX operating system functions for both software and hardware threads, ReconOS particularly caters to developers with a software background, because developers can use well-known mechanisms such as semaphores, mutexes, condition variables, and message queues for developing hybrid applications with threads running on the CPU and FPGA concurrently. Through the semantic integration of hardware accelerators into a standard operating system environment, ReconOS allows for rapid design space exploration, supports a structured application development process and improves the portability of applications between different reconfigurable computing systems.


Self-aware Compute Nodes

A. Agne, M. Happe, A. Lösch, C. Plessl, M. Platzner, in: Self-aware Computing Systems, Springer International Publishing, 2016, pp. 145-165

Many modern compute nodes are heterogeneous multi-cores that integrate several CPU cores with fixed function or reconfigurable hardware cores. Such systems need to adapt task scheduling and mapping to optimise for performance and energy under varying workloads and, increasingly important, for thermal and fault management and are thus relevant targets for self-aware computing. In this chapter, we take up the generic reference architecture for designing self-aware and self-expressive computing systems and refine it for heterogeneous multi-cores. We present ReconOS, an architecture, programming model and execution environment for heterogeneous multi-cores, and show how the components of the reference architecture can be implemented on top of ReconOS. In particular, the unique feature of dynamic partial reconfiguration supports self-expression through starting and terminating reconfigurable hardware cores. We detail a case study that runs two applications on an architecture with one CPU and 12 reconfigurable hardware cores and present self-expression strategies for adapting under performance, temperature and even conflicting constraints. The case study demonstrates that the reference architecture as a model for self-aware computing is highly useful as it allows us to structure and simplify the design process, which will be essential for designing complex future compute nodes. Furthermore, ReconOS is used as a base technology for flexible protocol stacks in Chapter 10, an approach for self-aware computing at the networking level.


Using Approximate Computing in Scientific Codes

M. Lass, T. Kühne, C. Plessl, in: Workshop on Approximate Computing (AC), 2016


Using Just-in-Time Code Generation for Transparent Resource Management in Heterogeneous Systems

H. Riebler, G.F. Vaz, C. Plessl, E.M.G. Trainiti, G.C. Durelli, C. Bolchini, in: Proc. HiPEAC Workshop on Reonfigurable Computing (WRC), 2016


Using Just-in-Time Code Generation for Transparent Resource Management in Heterogeneous Systems

H. Riebler, G.F. Vaz, C. Plessl, E.M.G.. Trainiti, G.C. Durelli, E. Del Sozzo, M.D.. Santambrogio, C. Bolchini, in: Proceedings of International Forum on Research and Technologies for Society and Industry (RTSI), IEEE, 2016, pp. 1-5

Hardware accelerators are becoming popular in academia and industry. To move one step further from the state-of-the-art multicore plus accelerator approaches, we present in this paper our innovative SAVEHSA architecture. It comprises of a heterogeneous hardware platform with three different high-end accelerators attached over PCIe (GPGPU, FPGA and Intel MIC). Such systems can process parallel workloads very efficiently whilst being more energy efficient than regular CPU systems. To leverage the heterogeneity, the workload has to be distributed among the computing units in a way that each unit is well-suited for the assigned task and executable code must be available. To tackle this problem we present two software components; the first can perform resource allocation at runtime while respecting system and application goals (in terms of throughput, energy, latency, etc.) and the second is able to analyze an application and generate executable code for an accelerator at runtime. We demonstrate the first proof-of-concept implementation of our framework on the heterogeneous platform, discuss different runtime policies and measure the introduced overheads.


2015

Aktuelles Schlagwort: Approximate Computing

C. Plessl, M. Platzner, P.J. Schreier, Informatik Spektrum (2015)(5), pp. 396-399

DOI


Easy-to-Use On-The-Fly Binary Program Acceleration on Many-Cores

M. Damschen, C. Plessl, in: Proceedings of the 5th International Workshop on Adaptive Self-tuning Computing Systems (ADAPT), 2015

This paper introduces Binary Acceleration At Runtime(BAAR), an easy-to-use on-the-fly binary acceleration mechanismwhich aims to tackle the problem of enabling existentsoftware to automatically utilize accelerators at runtime. BAARis based on the LLVM Compiler Infrastructure and has aclient-server architecture. The client runs the program to beaccelerated in an environment which allows program analysisand profiling. Program parts which are identified as suitable forthe available accelerator are exported and sent to the server.The server optimizes these program parts for the acceleratorand provides RPC execution for the client. The client transformsits program to utilize accelerated execution on the server foroffloaded program parts. We evaluate our work with a proofof-concept implementation of BAAR that uses an Intel XeonPhi 5110P as the acceleration target and performs automaticoffloading, parallelization and vectorization of suitable programparts. The practicality of BAAR for real-world examples is shownbased on a study of stencil codes. Our results show a speedup ofup to 4 without any developer-provided hints and 5.77 withhints over the same code compiled with the Intel Compiler atoptimization level O2 and running on an Intel Xeon E5-2670machine. Based on our insights gained during implementationand evaluation we outline future directions of research, e.g.,offloading more fine-granular program parts than functions, amore sophisticated communication mechanism or introducing onstack-replacement.


Exploring Tradeoffs between Specialized Kernels and a Reusable Overlay in a Stereo-Matching Case Study

T. Kenter, H. Schmitz, C. Plessl, International Journal of Reconfigurable Computing (IJRC) (2015), 2015, 859425

FPGAs are known to permit huge gains in performance and efficiency for suitable applications but still require reduced design efforts and shorter development cycles for wider adoption. In this work, we compare the resulting performance of two design concepts that in different ways promise such increased productivity. As common starting point, we employ a kernel-centric design approach, where computational hotspots in an application are identified and individually accelerated on FPGA. By means of a complex stereo matching application, we evaluate two fundamentally different design philosophies and approaches for implementing the required kernels on FPGAs. In the first implementation approach, we designed individually specialized data flow kernels in a spatial programming language for a Maxeler FPGA platform; in the alternative design approach, we target a vector coprocessor with large vector lengths, which is implemented as a form of programmable overlay on the application FPGAs of a Convey HC-1. We assess both approaches in terms of overall system performance, raw kernel performance, and performance relative to invested resources. After compensating for the effects of the underlying hardware platforms, the specialized dataflow kernels on the Maxeler platform are around 3x faster than kernels executing on the Convey vector coprocessor. In our concrete scenario, due to trade-offs between reconfiguration overheads and exposed parallelism, the advantage of specialized dataflow kernels is reduced to around 2.5x.


FELIX: a High-Throughput Network Approach for Interfacing to Front End Electronics for ATLAS Upgrades

J. Anderson, A. Borga, H. Boterenbrood, H. Chen, K. Chen, G. Drake, D. Francis, B. Gorini, F. Lanni, G. Lehmann Miotto, L. Levinson, J. Narevicius, C. Plessl, A. Roich, S. Ryu, F. Schreuder, J. Schumacher, W. Vandelli, J. Vermeulen, J. Zhang, Journal of Physics: Conference Series (2015), 664, 082050

The ATLAS experiment at CERN is planning full deployment of a new unified optical link technology for connecting detector front end electronics on the timescale of the LHC Run 4 (2025). It is estimated that roughly 8000 GBT (GigaBit Transceiver) links, with transfer rates up to 10.24 Gbps, will replace existing links used for readout, detector control and distribution of timing and trigger information. A new class of devices will be needed to interface many GBT links to the rest of the trigger, data-acquisition and detector control systems. In this paper FELIX (Front End LInk eXchange) is presented, a PC-based device to route data from and to multiple GBT links via a high-performance general purpose network capable of a total throughput up to O(20 Tbps). FELIX implies architectural changes to the ATLAS data acquisition system, such as the use of industry standard COTS components early in the DAQ chain. Additionally the design and implementation of a FELIX demonstration platform is presented and hardware and software aspects will be discussed.


Improving Packet Processing Performance in the ATLAS FELIX Project – Analysis and Optimization of a Memory-Bounded Algorithm

J. Schumacher, J. T. Anderson, A. Borga, H. Boterenbrood, H. Chen, K. Chen, G. Drake, D. Francis, B. Gorini, F. Lanni, G. Lehmann-Miotto, L. Levinson, J. Narevicius, C. Plessl, A. Roich, S. Ryu, F. P. Schreuder, W. Vandelli, J. Vermeulen, J. Zhang, in: Proc. Int. Conf. on Distributed Event-Based Systems (DEBS), ACM, 2015

DOI


Management and Scheduling of Accelerators for Heterogeneous High-Performance Computing

T. Beisel, Logos Verlag Berlin GmbH, 2015

The use of heterogeneous computing resources, such as graphics processing units or other specialized co-processors, has become widespread in recent years because of their performance and energy efficiency advantages. Operating system approaches that are limited to optimizing CPU usage are no longer sufficient for the efficient utilization of systems that comprise diverse resource types. Enabling task preemption on these architectures and migration of tasks between different resource types at run-time is not only key to improving the performance and energy consumption but also to enabling automatic scheduling methods for heterogeneous compute nodes. This thesis proposes novel techniques for run-time management of heterogeneous resources and enabling tasks to migrate between diverse hardware. It provides fundamental work towards future operating systems by discussing implications, limitations, and chances of the heterogeneity and introducing solutions for energy- and performance-efficient run-time systems. Scheduling methods to utilize heterogeneous systems by the use of a centralized scheduler are presented that show benefits over existing approaches in varying case studies.


Multiobjective Optimization of the Flow Around a Cylinder Using Model Order Reduction

S. Peitz, M. Dellnitz, PAMM (2015), 15(1), pp. 613-614

In this article an efficient numerical method to solve multiobjective optimization problems for fluid flow governed by the Navier Stokes equations is presented. In order to decrease the computational effort, a reduced order model is introduced using Proper Orthogonal Decomposition and a corresponding Galerkin Projection. A global, derivative free multiobjective optimization algorithm is applied to compute the Pareto set (i.e. the set of optimal compromises) for the concurrent objectives minimization of flow field fluctuations and control cost. The method is illustrated for a 2D flow around a cylinder at Re = 100.


Self-Aware and Self-Expressive Systems – Guest Editor's Introduction

J. Torresen, C. Plessl, X. Yao, IEEE Computer (2015), 48(7), pp. 18-20

DOI


Simulative Ultraschall-Untersuchung von Pitch-Catch-Messanordnungen für große zylindrische Stahl-Prüflinge und gradientenbasierte Bildgebung

S. Hegler, C. Statz, M. Mütze, H. Mooshofer, M. Goldammer, K. Fendt, S. Schwarzer, K. Feldhoff, M. Flehmig, U. Markwardt, W. E. Nagel, M. Schütte, A. Walther, M. Meinel, A. Basermann, D. Plettemeier, tm - Technisches Messen (2015), 82(9), pp. 440-450

Große zylindrische Stahlprüflinge werden mittels der Methode der finiten Differenzen im Zeitbereich (engl. finite differences in time domain, FDTD) simulativ untersucht. Dabei werden Pitch-Catch-Messanordnungen verwendet. Es werden zwei Bildgebungsansätze vorgestellt: ersterer basiert auf dem Imaging Principle nach Claerbout, letzterer basiert auf gradientenbasierter Optimierung eines Zielfunktionals.


Transparent offloading of computational hotspots from binary code to Xeon Phi

M. Damschen, H. Riebler, G.F. Vaz, C. Plessl, in: Proceedings of the 2015 Conference on Design, Automation and Test in Europe (DATE), EDA Consortium / IEEE, 2015, pp. 1078-1083

In this paper, we study how binary applications can be transparently accelerated with novel heterogeneous computing resources without requiring any manual porting or developer-provided hints. Our work is based on Binary Acceleration At Runtime (BAAR), our previously introduced binary acceleration mechanism that uses the LLVM Compiler Infrastructure. BAAR is designed as a client-server architecture. The client runs the program to be accelerated in an environment, which allows program analysis and profiling and identifies and extracts suitable program parts to be offloaded. The server compiles and optimizes these offloaded program parts for the accelerator and offers access to these functions to the client with a remote procedure call (RPC) interface. Our previous work proved the feasibility of our approach, but also showed that communication time and overheads limit the granularity of functions that can be meaningfully offloaded. In this work, we motivate the importance of a lightweight, high-performance communication between server and client and present a communication mechanism based on the Message Passing Interface (MPI). We evaluate our approach by using an Intel Xeon Phi 5110P as the acceleration target and show that the communication overhead can be reduced from 40% to 10%, thus enabling even small hotspots to benefit from offloading to an accelerator.


2014

Accelerating Finite Difference Time Domain Simulations with Reconfigurable Dataflow Computers

H. Giefers, C. Plessl, J. Förstner, ACM SIGARCH Computer Architecture News (2014), 41(5), pp. 65-70

DOI


Deferring Accelerator Offloading Decisions to Application Runtime

G.F. Vaz, H. Riebler, T. Kenter, C. Plessl, in: Proceedings of the International Conference on ReConFigurable Computing and FPGAs (ReConFig), IEEE, 2014, pp. 1-8

Reconfigurable architectures provide an opportunityto accelerate a wide range of applications, frequentlyby exploiting data-parallelism, where the same operations arehomogeneously executed on a (large) set of data. However, whenthe sequential code is executed on a host CPU and only dataparallelloops are executed on an FPGA coprocessor, a sufficientlylarge number of loop iterations (trip counts) is required, such thatthe control- and data-transfer overheads to the coprocessor canbe amortized. However, the trip count of large data-parallel loopsis frequently not known at compile time, but only at runtime justbefore entering a loop. Therefore, we propose to generate codeboth for the CPU and the coprocessor, and to defer the decisionwhere to execute the appropriate code to the runtime of theapplication when the trip count of the loop can be determinedjust at runtime. We demonstrate how an LLVM compiler basedtoolflow can automatically insert appropriate decision blocks intothe application code. Analyzing popular benchmark suites, weshow that this kind of runtime decisions is often applicable. Thepractical feasibility of our approach is demonstrated by a toolflowthat automatically identifies loops suitable for vectorization andgenerates code for the FPGA coprocessor of a Convey HC-1. Thetoolflow adds decisions based on a comparison of the runtimecomputedtrip counts to thresholds for specific loops and alsoincludes support to move just the required data to the coprocessor.We evaluate the integrated toolflow with characteristic loopsexecuted on different input data sizes.


Kernel-Centric Acceleration of High Accuracy Stereo-Matching

T. Kenter, H. Schmitz, C. Plessl, in: Proceedings of the International Conference on ReConFigurable Computing and FPGAs (ReConFig), IEEE, 2014, pp. 1-8

Stereo-matching algorithms recently received a lot of attention from the FPGA acceleration community. Presented solutions range from simple, very resource efficient systems with modest matching quality for small embedded systems to sophisticated algorithms with several processing steps, implemented on big FPGAs. In order to achieve high throughput, most implementations strongly focus on pipelining and data reuse between different computation steps. This approach leads to high efficiency, but limits the supported computation patterns and due the high integration of the implementation, adaptions to the algorithm are difficult. In this work, we present a stereo-matching implementation, that starts by offloading individual kernels from the CPU to the FPGA. Between subsequent compute steps on the FPGA, data is stored off-chip in on-board memory of the FPGA accelerator card. This enables us to accelerate the AD-census algorithm with cross-based aggregation and scanline optimization for the first time without algorithmic changes and for up to full HD image dimensions. Analyzing throughput and bandwidth requirements, we outline some trade-offs that are involved with this approach, compared to tighter integration of more kernel loops into one design.


Numerical Simulation of the Damping Behavior of Particle-Filled Hollow Spheres

T. Steinle, J. Vrabec, A. Walther, in: Proc. Modeling, Simulation and Optimization of Complex Processes (HPSC), Springer International Publishing, 2014, pp. 233-243

In light of an increasing awareness of environmental challenges, extensive research is underway to develop new light-weight materials. A problem arising with these materials is their increased response to vibration. This can be solved using a new composite material that contains embedded hollow spheres that are partially filled with particles. Progress on the adaptation of molecular dynamics towards a particle-based numerical simulation of this material is reported. This includes the treatment of specific boundary conditions and the adaption of the force computation. First results are presented that showcase the damping properties of such particle-filled spheres in a bouncing experiment.


On Semeai Detection in Monte-Carlo Go

T. Graf, L. Schaefers, M. Platzner, in: Proc. Conf. on Computers and Games (CG), Springer, 2014, pp. 14-25

DOI


Partitioning and Vectorizing Binary Applications for a Reconfigurable Vector Computer

T. Kenter, G.F. Vaz, C. Plessl, in: Proceedings of the International Symposium on Reconfigurable Computing: Architectures, Tools, and Applications (ARC), Springer International Publishing, 2014, pp. 144-155

In order to leverage the use of reconfigurable architectures in general-purpose computing, quick and automated methods to find suitable accelerator designs are required. We tackle this challenge in both regards. In order to avoid long synthesis times, we target a vector copro- cessor, implemented on the FPGAs of a Convey HC-1. Previous studies showed that existing tools were not able to accelerate a real-world application with low effort. We present a toolflow to automatically identify suitable loops for vectorization, generate a corresponding hardware/software bipartition, and generate coprocessor code. Where applicable, we leverage outer-loop vectorization. We evaluate our tools with a set of characteristic loops, systematically analyzing different dependency and data layout properties.


ReconOS - An Operating System Approach for Reconfigurable Computing

A. Agne, M. Happe, A. Keller, E. Lübbers, B. Plattner, M. Platzner, C. Plessl, IEEE Micro (2014), 34(1), pp. 60-71

The ReconOS operating system for reconfigurable computing offers a unified multi-threaded programming model and operating system services for threads executing in software and threads mapped to reconfigurable hardware. The operating system interface allows hardware threads to interact with software threads using well-known mechanisms such as semaphores, mutexes, condition variables, and message queues. By semantically integrating hardware accelerators into a standard operating system environment, ReconOS allows for rapid design space exploration, supports a structured application development process and improves the portability of applications


Reconstructing AES Key Schedules from Decayed Memory with FPGAs

H. Riebler, T. Kenter, C. Plessl, C. Sorge, in: Proceedings of Field-Programmable Custom Computing Machines (FCCM), IEEE, 2014, pp. 222-229

In this paper, we study how AES key schedules can be reconstructed from decayed memory. This operation is a crucial and time consuming operation when trying to break encryption systems with cold-boot attacks. In software, the reconstruction of the AES master key can be performed using a recursive, branch-and-bound tree-search algorithm that exploits redundancies in the key schedule for constraining the search space. In this work, we investigate how this branch-and-bound algorithm can be accelerated with FPGAs. We translated the recursive search procedure to a state machine with an explicit stack for each recursion level and create optimized datapaths to accelerate in particular the processing of the most frequently accessed tree levels. We support two different decay models, of which especially the more realistic non-idealized asymmetric decay model causes very high runtimes in software. Our implementation on a Maxeler dataflow computing system outperforms a software implementation for this model by up to 27x, which makes cold-boot attacks against AES practical even for high error rates.


Runtime Resource Management in Heterogeneous System Architectures: The SAVE Approach

G. C. Durelli, M. Pogliani, A. Miele, C. Plessl, H. Riebler, G.F. Vaz, M. D. Santambrogio, C. Bolchini, in: Proc. Int. Symp. on Parallel and Distributed Processing with Applications (ISPA), IEEE, 2014, pp. 142-149

DOI


SAVE: Towards efficient resource management in heterogeneous system architectures

G. C. Durelli, M. Copolla, K. Djafarian, G. Koranaros, A. Miele, M. Paolino, O. Pell, C. Plessl, M. D. Santambrogio, C. Bolchini, in: Proc. Int. Conf. on Reconfigurable Computing: Architectures, Tools and Applications (ARC), Springer, 2014

DOI


Self-awareness as a Model for Designing and Operating Heterogeneous Multicores

A. Agne, M. Happe, A. Lösch, C. Plessl, M. Platzner, ACM Transactions on Reconfigurable Technology and Systems (TRETS) (2014), 7(2), 13

Self-aware computing is a paradigm for structuring and simplifying the design and operation of computing systems that face unprecedented levels of system dynamics and thus require novel forms of adaptivity. The generality of the paradigm makes it applicable to many types of computing systems and, previously, researchers started to introduce concepts of self-awareness to multicore architectures. In our work we build on a recent reference architectural framework as a model for self-aware computing and instantiate it for an FPGA-based heterogeneous multicore running the ReconOS reconfigurable architecture and operating system. After presenting the model for self-aware computing and ReconOS, we demonstrate with a case study how a multicore application built on the principle of self-awareness, autonomously adapts to changes in the workload and system state. Our work shows that the reference architectural framework as a model for self-aware computing can be practically applied and allows us to structure and simplify the design process, which is essential for designing complex future computing systems.


Seven Recipes for Setting Your FPGA on Fire – A Cookbook on Heat Generators

A. Agne, H. Hangmann, M. Happe, M. Platzner, C. Plessl, Microprocessors and Microsystems (2014), 38(8, Part B), pp. 911-919

Due to the continuously shrinking device structures and increasing densities of FPGAs, thermal aspects have become the new focus for many research projects over the last years. Most researchers rely on temperature simulations to evaluate their novel thermal management techniques. However, these temperature simulations require a high computational effort if a detailed thermal model is used and their accuracies are often unclear. In contrast to simulations, the use of synthetic heat sources allows for experimental evaluation of temperature management methods. In this paper we investigate the creation of significant rises in temperature on modern FPGAs to enable future evaluation of thermal management techniques based on experiments. To that end, we have developed seven different heat-generating cores that use different subsets of FPGA resources. Our experimental results show that, according to external temperature probes connected to the FPGA’s heat sink, we can increase the temperature by an average of 81 !C. This corresponds to an average increase of 156.3 !C as measured by the built-in thermal diodes of our Virtex-5 FPGAs in less than 30 min by only utilizing about 21 percent of the slices.


Verschiebungen an der Grenze zwischen Hardware und Software

M. Platzner, C. Plessl, in: Logiken strukturbildender Prozesse: Automatismen, Wilhelm Fink, 2014, pp. 123-144

Im Bereich der Computersysteme ist die Festlegung der Grenze zwischen Hardware und Software eine zentrale Problemstellung. Diese Grenze hat in den letzten Jahrzehnten nicht nur die Entwicklung von Computersystemen bestimmt, sondern auch die Strukturierung der Ausbildung in den Computerwissenschaften beeinflusst und sogar zur Entstehung von neuen Forschungsrichtungen gef{\"u}hrt. In diesem Beitrag besch{\"a}ftigen wir uns mit Verschiebungen an der Grenze zwischen Hardware und Software und diskutieren insgesamt drei qualitativ unterschiedliche Formen solcher Verschiebungen. Wir beginnen mit der Entwicklung von Computersystemen im letzten Jahrhundert und der Entstehung dieser Grenze, die Hardware und Software erst als eigenst{\"a}ndige Produkte differenziert. Dann widmen wir uns der Frage, welche Funktionen in einem Computersystem besser in Hardware und welche besser in Software realisiert werden sollten, eine Fragestellung die zu Beginn der 90er-Jahre zur Bildung einer eigenen Forschungsrichtung, dem sogenannten Hardware/Software Co-design, gef{\"u}hrt hat. Im Hardware/Software Co-design findet eine Verschiebung von Funktionen an der Grenze zwischen Hardware und Software w{\"a}hrend der Entwicklung eines Produktes statt, um Produkteigenschaften zu optimieren. Im fertig entwickelten und eingesetzten Produkt hingegen k{\"o}nnen wir dann eine feste Grenze zwischen Hardware und Software beobachten. Im dritten Teil dieses Beitrags stellen wir mit selbst-adaptiven Systemen eine hochaktuelle Forschungsrichtung vor. In unserem Kontext bedeutet Selbstadaption, dass ein System Verschiebungen von Funktionen an der Grenze zwischen Hardware und Software autonom w{\"a}hrend der Betriebszeit vornimmt. Solche Systeme beruhen auf rekonfigurierbarer Hardware, einer relativ neuen Technologie mit der die Hardware eines Computers w{\"a}hrend der Laufzeit ver{\"a}ndert werden kann. Diese Technologie f{\"u}hrt zu einer durchl{\"a}ssigen Grenze zwischen Hardware und Software bzw. l{\"o}st sie die herk{\"o}mmliche Vorstellung einer festen Hardware und einer flexiblen Software damit auf.


2013

Advanced Data Deduplication Techniques and Their Application

D. Meister, Johannes Gutenberg-Universität Mainz, 2013



Distributing Storage in Cloud Environments

P. Berenbrink, A. Brinkmann, T. Friedetzky, D. Meister, L. Nagel, in: Proc. Int. Symp. on Parallel and Distributed Processing Workshops (IPDPSW), IEEE, 2013

DOI


File Recipe Compression in Data Deduplication Systems

D. Meister, A. Brinkmann, T. Süß, in: Proc. USENIX Conference on File and Storage Technologies (FAST), USENIX Association, 2013, pp. 175-182


FPGA Implementation of a Second-Order Convolutive Blind Signal Separation Algorithm

S. Kasap, S. Redif, in: Proc. IEEE Signal Processing and Communications Conf. (SUI), IEEE, 2013

DOI


FPGA-accelerated Key Search for Cold-Boot Attacks against AES

H. Riebler, T. Kenter, C. Sorge, C. Plessl, in: Proceedings of the International Conference on Field-Programmable Technology (FPT), IEEE, 2013, pp. 386-389

Cold-boot attacks exploit the fact that DRAM contents are not immediately lost when a PC is powered off. Instead the contents decay rather slowly, in particular if the DRAM chips are cooled to low temperatures. This effect opens an attack vector on cryptographic applications that keep decrypted keys in DRAM. An attacker with access to the target computer can reboot it or remove the RAM modules and quickly copy the RAM contents to non-volatile memory. By exploiting the known cryptographic structure of the cipher and layout of the key data in memory, in our application an AES key schedule with redundancy, the resulting memory image can be searched for sections that could correspond to decayed cryptographic keys; then, the attacker can attempt to reconstruct the original key. However, the runtime of these algorithms grows rapidly with increasing memory image size, error rate and complexity of the bit error model, which limits the practicability of the approach.In this work, we study how the algorithm for key search can be accelerated with custom computing machines. We present an FPGA-based architecture on a Maxeler dataflow computing system that outperforms a software implementation up to 205x, which significantly improves the practicability of cold-attacks against AES.


MCD: Overcoming the Data Download Bottleneck in Data Centers

J. Kaiser, D. Meister, V. Gottfried, A. Brinkmann, in: Proc. IEEE Int. Conf. on Networking, Architecture and Storage (NAS), IEEE Computer Society, 2013, pp. 88-97

DOI


Novel Field-Programmable Gate Array Architecture for Computing the Eigenvalue Decomposition of Para-Hermitian Polynomial Matrices

S. Kasap, S. Redif, IEEE Trans. on Very Large Scale Integration (VLSI) Systems (2013), 22(3), pp. 522-536

DOI


On-The-Fly Computing: A Novel Paradigm for Individualized IT Services

M. Happe, P. Kling, C. Plessl, M. Platzner, F. Meyer auf der Heide, in: Proceedings of the 9th IEEE Workshop on Software Technology for Future embedded and Ubiquitous Systems (SEUS), IEEE, 2013

In this paper we introduce “On-The-Fly Computing”, our vision of future IT services that will be provided by assembling modular software components available on world-wide markets. After suitable components have been found, they are automatically integrated, configured and brought to execution in an On-The-Fly Compute Center. We envision that these future compute centers will continue to leverage three current trends in large scale computing which are an increasing amount of parallel processing, a trend to use heterogeneous computing resources, and—in the light of rising energy cost—energy-efficiency as a primary goal in the design and operation of computing systems. In this paper, we point out three research challenges and our current work in these areas.


Parallel Macro Pipelining on the Intel SCC Many-Core Computer

T. Suess, A. Schoenrock, S. Meisner, C. Plessl, in: Proc. Int. Symp. on Parallel and Distributed Processing Workshops (IPDPSW), IEEE Computer Society, 2013, pp. 64-73

DOI


2012

A Data Driven Science Gateway for Computational Workflows

R. Grunzke, G. Birkenheuer, D. Blunk, S. Breuers, A. Brinkmann, S. Gesing, S. Herres-Pawlis, O. Kohlbacher, J. Krüger, M. Kruse, R. Müller-Pfefferkorn, P. Schäfer, B. Schuller, T. Steinke, A. Zink, in: Proc. UNICORE Summit, 2012


A Science Gateway Getting Ready for Serving the International Molecular Simulation Community

S. Gesing, S. Herres-Pawlis, G. Birkenheuer, A. Brinkmann, R. Grunzke, P. Kacsuk, O. Kohlbacher, M. Kozlovszky, J. Krüger, R. Müller-Pfefferkorn, P. Schäfer, T. Steinke, in: Proceedings of Science, 2012


A Single Sign-On Infrastructure for Science Gateways on a Use Case for Structural Bioinformatics

S. Gesing, R. Grunzke, J. Krüger, G. Birkenheuer, M. Wewior, P. Schäfer, B. Schuller, J. Schuster, S. Herres-Pawlis, S. Breuers, Balaskó, M. Kozlovszky, A. Szikszay Fabri, L. Packschies, P. Kacsuk, D. Blunk, T. Steinke, A. Brinkmann, G. Fels, R. Müller-Pfefferkorn, R. Jäkel, O. Kohlbacher, Journal of Grid Computing (2012), 10(4), pp. 769-790

DOI


A Study on Data Deduplication in HPC Storage Systems

D. Meister, J. Kaiser, A. Brinkmann, M. Kuhn, J. Kunkel, T. Cortes, in: Proc. Int. Conf. on Supercomputing (SC), IEEE Computer Society, 2012, pp. 7:1-7:11

DOI


Comparison of Bayesian Move Prediction Systems for Computer Go

M. Wistuba, L. Schaefers, M. Platzner, in: Proc. IEEE Conf. on Computational Intelligence and Games (CIG), IEEE, 2012, pp. 91-99

DOI


Convey Vector Personalities – FPGA Acceleration with an OpenMP-like Effort?

B. Meyer, J. Schumacher, C. Plessl, J. Förstner, in: Proc. Int. Conf. on Field Programmable Logic and Applications (FPL), IEEE, 2012, pp. 189-196

Although the benefits of FPGAs for accelerating scientific codes are widely acknowledged, the use of FPGA accelerators in scientific computing is not widespread because reaping these benefits requires knowledge of hardware design methods and tools that is typically not available with domain scientists. A promising but hardly investigated approach is to develop tool flows that keep the common languages for scientific code (C,C++, and Fortran) and allow the developer to augment the source code with OpenMPlike directives for instructing the compiler which parts of the application shall be offloaded the FPGA accelerator. In this work we study whether the promise of effective FPGA acceleration with an OpenMP-like programming effort can actually be held. Our target system is the Convey HC-1 reconfigurable computer for which an OpenMP-like programming environment exists. As case study we use an application from computational nanophotonics. Our results show that a developer without previous FPGA experience could create an FPGA-accelerated application that is competitive to an optimized OpenMP-parallelized CPU version running on a two socket quad-core server. Finally, we discuss our experiences with this tool flow and the Convey HC-1 from a productivity and economic point of view.


Cost-aware and SLO Fulfilling Software as a Service

O. Niehörster, J. Simon, A. Brinkmann, A. Keller, J. Krüger, Journal of Grid Computing (2012), 10(3), pp. 553-577

Virtualization technology makes data centers more dynamic and easier to administrate. Today, cloud providers offer customers access to complex applications running on virtualized hardware. Nevertheless, big virtualized data centers become stochastic environments and the simplification on the user side leads to many challenges for the provider. He has to find cost-efficient configurations and has to deal with dynamic environments to ensure service level objectives (SLOs). We introduce a software solution that reduces the degree of human intervention to manage clouds. It is designed as a multi-agent system (MAS) and placed on top of the Infrastructure as a Service (IaaS) layer. Worker agents allocate resources, configure applications, check the feasibility of requests, and generate cost estimates. They are equipped with application specific knowledge allowing it to estimate the type and number of necessary resources. During runtime, a worker agent monitors the job and adapts its resources to ensure the specified quality of service—even in noisy clouds where the job instances are influenced by other jobs. They interact with a scheduler agent, which takes care of limited resources and does a cost-aware scheduling by assigning jobs to times with low costs. The whole architecture is self-optimizing and able to use public or private clouds. Building a private cloud needs to face the challenge to find a mapping of virtual machines (VMs) to hosts. We present a rule-based mapping algorithm for VMs. It offers an interface where policies can be defined and combined in a generic way. The algorithm performs the initial mapping at request time as well as a remapping during runtime. It deals with policy and infrastructure changes. An energy-aware scheduler and the availability of cheap resources provided by a spot market are analyzed. We evaluated our approach by building up an SaaS stack, which assigns resources in consideration of an energy function and that ensures SLOs of two different applications, a brokerage system and a high-performance computing software. Experiments were done on a real cloud system and by simulations.


Design of an exact data deduplication cluster

J. Kaiser, D. Meister, A. Brinkmann, S. Effert, in: Proc. Symp. on Mass Storage Systems and Technologies (MSST), IEEE, 2012, pp. 1-12

DOI


Eight Ways to put your FPGA on Fire – A Systematic Study of Heat Generators

M. Happe, H. Hangmann, A. Agne, C. Plessl, in: Proceedings of the International Conference on Reconfigurable Computing and FPGAs (ReConFig), IEEE, 2012, pp. 1-8

Due to the continuously shrinking device structures and increasing densities of FPGAs, thermal aspects have become the new focus for many research projects over the last years. Most researchers rely on temperature simulations to evaluate their novel thermal management techniques. However, the accuracy of the simulations is to some extent questionable and they require a high computational effort if a detailed thermal model is used.For experimental evaluation of real-world temperature management methods, often synthetic heat sources are employed. Therefore, in this paper we investigated the question if we can create significant rises in temperature on modern FPGAs to enable future evaluation of thermal management techniques based on experiments in contrast to simulations. Therefore, we have developed eight different heat-generating cores that use different subsets of the FPGA resources. Our experimental results show that, according to the built-in thermal diode of our Xilinx Virtex-5 FPGA, we can increase the chip temperature by 134 degree C in less than 12 minutes by only utilizing about 21% of the slices.


ESB: Ext2 Split Block Device

J. Kaiser, D. Meister, T. Hartung, A. Brinkmann, in: Proc. IEEE Int. Conf. on Parallel and Distributed Systems (ICPADS), IEEE, 2012, pp. 181-188

DOI


Exploration of Ring Oscillator Design Space for Temperature Measurements on FPGAs

C. Rüthing, M. Happe, A. Agne, C. Plessl, in: Proceedings of the International Conference on Field Programmable Logic and Applications (FPL), IEEE, 2012, pp. 559-562

While numerous publications have presented ring oscillator designs for temperature measurements a detailed study of the ring oscillator's design space is still missing. In this work, we introduce metrics for comparing the performance and area efficiency of ring oscillators and a methodology for determining these metrics. As a result, we present a systematic study of the design space for ring oscillators for a Xilinx Virtex-5 platform FPGA.


FPGA implementation of a second-order convolutive blind signal separation algorithm

S. Kasap, S. Redif, in: Int. Architecture and Engineering Symp. (ARCHENG), 2012


FPGA-based design and implementation of an approximate polynomial matrix EVD algorithm

S. Kasap, S. Redif, in: Proc. Int. Conf. on Field Programmable Technology (ICFPT), IEEE Computer Society, 2012, pp. 135-140

DOI


Generic User Management for Science Gateways via Virtual Organizations

T. Schlemmer, R. Grunzke, S. Gesing, J. Krüger, G. Birkenheuer, R. Müller-Pfefferkorn, O. Kohlbacher, in: Proc. EGI Technical Forum, 2012


Hardware/Software Platform for Self-aware Compute Nodes

M. Happe, A. Agne, C. Plessl, M. Platzner, in: Proceedings of the Workshop on Self-Awareness in Reconfigurable Computing Systems (SRCS), 2012, pp. 8-9

Today's design and operation principles and methods do not scale well with future reconfigurable computing systems due to an increased complexity in system architectures and applications, run-time dynamics and corresponding requirements. Hence, novel design and operation principles and methods are needed that possibly break drastically with the static ones we have built into our systems and the fixed abstraction layers we have cherished over the last decades. Thus, we propose a HW/SW platform that collects and maintains information about its state and progress which enables the system to reason about its behavior (self-awareness) and utilizes its knowledge to effectively and autonomously adapt its behavior to changing requirements (self-expression).To enable self-awareness, our compute nodes collect information using a variety of sensors, i.e. performance counters and thermal diodes, and use internal self-awareness models that process these information. For self-awareness, on-line learning is crucial such that the node learns and continuously updates its models at run-time to react to changing conditions. To enable self-expression, we break with the classic design-time abstraction layers of hardware, operating system and software. In contrast, our system is able to vertically migrate functionalities between the layers at run-time to exploit trade-offs between abstraction and optimization.This paper presents a heterogeneous multi-core architecture, that enables self-awareness and self-expression, an operating system for our proposed hardware/software platform and a novel self-expression method.


IMORC: An Infrastructure and Architecture Template for Implementing High-Performance Reconfigurable FPGA Accelerators

T. Schumacher, C. Plessl, M. Platzner, Microprocessors and Microsystems (2012), 36(2), pp. 110-126

DOI


On the Feasibility and Limitations of Just-In-Time Instruction Set Extension for FPGA-based Reconfigurable Processors

M. Grad, C. Plessl, Int. Journal of Reconfigurable Computing (IJRC) (2012)

DOI


One Phase Commit: A Low Overhead Atomic Commitment Protocol for Scalable Metadata Services

G. Congiu, M. Grawinkel, S. Narasimhamurthy, A. Brinkmann, in: Proc. Workshop on Interfaces and Architectures for Scientific Data Storage (IASDS), IEEE, 2012, pp. 16-24

DOI


Parallel algorithm for computation of second-order sequential best rotations

S. Redif, S. Kasap, Int. Journal of Electronics (2012), 100(12), pp. 1646-1651

DOI


Parallel Processor Design and Implementation for Molecular Dynamics Simulations on a FPGA Parallel Computer

S. Kasap, K. Benkrid, Journal of Computers (2012), 7(6), pp. 1312-1328


Pragma based parallelization - Trading hardware efficiency for ease of use?

T. Kenter, C. Plessl, H. Schmitz, in: Proceedings of the International Conference on ReConFigurable Computing and FPGAs (ReConFig), IEEE, 2012, pp. 1-8

One major obstacle for a wide spread FPGA usage in general-purpose computing is the development tool flow that requires much higher effort than for pure software solutions. Convey Computer promises a solution to this problem for their HC-1 platform, where the FPGAs are configured to run as a vector processor and the software source code can be annotated with pragmas that guide an automated vectorization process. We investigate this approach for a stereo matching algorithm that has abundant parallelism and a number of different computational patterns. We note that for this case study the automated vectorization in its current state doesn’t hold its productivity promise. However, we also show that using the Vector Personality can yield a significant speedups compared to CPU implementations in two of three investigated phases of the algorithm. Those speedups don’t match custom FPGA implementations, but can come with much reduced development effort.


Programming and Scheduling Model for Supporting Heterogeneous Accelerators in Linux

T. Beisel, T. Wiersema, C. Plessl, A. Brinkmann, in: Proc. Workshop on Computer Architecture and Operating System Co-design (CAOS), 2012


STIR: Software for Tomographic Image Reconstruction Release 2

K. Thielemans, C. Tsoumpas, S. Mustafovic, T. Beisel, P. Aguiar, N. Dikaios, M. W Jacobson, Physics in Medicine and Biology (2012), 57(4), pp. 867-883

DOI


The MoSGrid Community From National to International Scale

S. Gesing, S. Herres-Pawlis, G. Birkenheuer, A. Brinkmann, R. Grunzke, P. Kacsuk, O. Kohlbacher, M. Kozlovszky, J. Krüger, R. Müller-Pfefferkorn, P. Schäfer, T. Steinke, in: Proc. EGI Community Forum, 2012


Towards Dynamic Scripted pNFS Layouts

M. Grawinkel, T. Süß, G. Best, I. Popov, A. Brinkmann, in: Proc. Parallel Data Storage Workshop (PDSW), IEEE, 2012, pp. 13-17

DOI


Turning control flow graphs into function calls: Code generation for heterogeneous architectures

P. Barrio, C. Carreras, R. Sierra, T. Kenter, C. Plessl, in: Proceedings of the International Conference on High Performance Computing and Simulation (HPCS), IEEE, 2012, pp. 559-565

Heterogeneous machines are gaining momentum in the High Performance Computing field, due to the theoretical speedups and power consumption. In practice, while some applications meet the performance expectations, heterogeneous architectures still require a tremendous effort from the application developers. This work presents a code generation method to port codes into heterogeneous platforms, based on transformations of the control flow into function calls. The results show that the cost of the function-call mechanism is affordable for the tested HPC kernels. The complete toolchain, based on the LLVM compiler infrastructure, is fully automated once the sequential specification is provided.


Workflow-enhanced conformational analysis of guanidine zinc complexes via a science gateway

S. Herres-Pawlis, G. Birkenheuer, A. Brinkmann, S. Gesing, R. Grunzke, R. Jäkel, O. Kohlbacher, J. Krüger, I. Dos Santos Vieira, Studies in Health Technology and Informatics (2012), 175, pp. 142-151

DOI


2011

A Science Gateway for Molecular Simulations

S. Gesing, P. Kacsuk, M. Kozlovszky, G. Birkenheuer, D. Blunk, S. Breuers, A. Brinkmann, G. Fels, R. Grunzke, S. Herres-Pawlis, J. Krüger, L. Packschies, R. Müller-Pfefferkorn, P. Schäfer, T. Steinke, A. Szikszay Fabri, K. Warzecha, M. Wewior, O. Kohlbacher, in: Proc. EGI User Forum, 2011, pp. 94-95


An Energy-Aware SaaS Stack

O. Niehörster, A. Keller, A. Brinkmann, in: Proc. Int. Meeting of the IEEE Int. Symp. on Modeling, Analysis and Simulation of Computer and Telecommunication Systems (MASCOTS), 2011

We present a multi-agent system on top of the IaaS layer consisting of a scheduler agent and multiple worker agents. Each job is controlled by an autonomous worker agent, which is equipped with application specific knowledge (e.g., performance functions) allowing it to estimate the type and number of necessary resources. During runtime, the worker agent monitors the job and adapts its resources to ensure the specified quality of service - even in noisy clouds where the job instances are influenced by other jobs. All worker agents interact with the scheduler agent, which takes care of limited resources and does a cost-aware scheduling by assigning jobs to times with low energy costs. The whole architecture is self-optimizing and able to use public or private clouds.


Autonomic Resource Management Handling Delayed Configuration Effects

O. Niehörster, A. Brinkmann, in: Proc. IEEE Int. Conf. on Cloud Computing Technology and Science (CloudCom), IEEE Computer Society, 2011, pp. 138-145

DOI


Autonomic Resource Management with Support Vector Machines

O. Niehörster, J. Simon, A. Brinkmann, A. Krieger, in: Proc. IEEE/ACM Int. Conf. on Grid Computing (GRID), IEEE Computer Society, 2011, pp. 157-164

DOI


Cooperative multitasking for heterogeneous accelerators in the Linux Completely Fair Scheduler

T. Beisel, T. Wiersema, C. Plessl, A. Brinkmann, in: Proc. Int. Conf. on Application-Specific Systems, Architectures, and Processors (ASAP), IEEE Computer Society, 2011, pp. 223-226

DOI


Estimation and Partitioning for CPU-Accelerator Architectures

T. Kenter, C. Plessl, M. Platzner, M. Kauschke, in: Intel European Research and Innovation Conference, 2011


Evaluation of Applied Intra-Disk Redundancy Schemes to Improve Single Disk Reliability

M. Grawinkel, T. Schäfer, A. Brinkmann, J. Hagemeyer, M. Porrmann, in: Proc. Int. Symp. on Modeling, Analysis and Simulation of Computer and Telecommunication Systems (MASCOTS), IEEE Computer Society, 2011, pp. 297-306

DOI


FPGA Acceleration of Communication-bound Streaming Applications: Architecture Modeling and a 3D Image Compositing Case Study

T. Schumacher, T. Süß, C. Plessl, M. Platzner, Int. Journal of Recon- figurable Computing (IJRC) (2011)

DOI


Granular Security for a Science Gateway in Structural Bioinformatics

S. Gesing, R. Grunzke, Balaskó, G. Birkenheuer, D. Blunk, S. Breuers, A. Brinkmann, G. Fels, S. Herres-Pawlis, P. Kacsuk, M. Kozlovszky, J. Krüger, L. Packschies, P. Schäfer, B. Schuller, J. Schuster, T. Steinke, A. Szikszay Fabri, M. Wewior, R. Müller-Pfefferkorn, O. Kohlbacher, in: Proc. Int. Workshop on Scientific Gateways (IWSG), Consorzio COMETA, 2011


Hardware Virtualization on Dynamically Reconfigurable Embedded Processors

C. Plessl, M. Platzner, in: Reconfigurable Embedded Control Systems: Applications for Flexibility and Agility, IGI Global, 2011

DOI


Infrastructure Federation Through Virtualized Delegation of Resources and Services

G. Birkenheuer, A. Brinkmann, M. Högqvist, A. Papaspyrou, B. Schott, D. Sommerfeld, W. Ziegler, Journal of Grid Computing (2011), 9(3), pp. 355-377

DOI


Just-in-time Instruction Set Extension – Feasibility and Limitations for an FPGA-based Reconfigurable ASIP Architecture

M. Grad, C. Plessl, in: Proc. Reconfigurable Architectures Workshop (RAW), IEEE Computer Society, 2011, pp. 278-285

DOI


Lonestar: An Energy-Aware Disk Based Long-Term Archival Storage System

M. Grawinkel, M. Pargmann, H. Dömer, A. Brinkmann, in: Proc. IEEE Int. Conf. on Parallel and Distributed Systems (ICPADS), IEEE, 2011, pp. 380-387

DOI


Measuring and Predicting Temperature Distributions on FPGAs at Run-Time

M. Happe, A. Agne, C. Plessl, in: Proceedings of the 2011 International Conference on Reconfigurable Computing and FPGAs (ReConFig), IEEE, 2011, pp. 55-60

In the next decades, hybrid multi-cores will be the predominant architecture for reconfigurable FPGA-based systems. Temperature-aware thread mapping strategies are key for providing dependability in such systems. These strategies rely on measuring the temperature distribution and redicting the thermal behavior of the system when there are changes to the hardware and software running on the FPGA. While there are a number of tools that use thermal models to predict temperature distributions at design time, these tools lack the flexibility to autonomously adjust to changing FPGA configurations. To address this problem we propose a temperature-aware system that empowers FPGA-based reconfigurable multi-cores to autonomously predict the on-chip temperature distribution for pro-active thread remapping. Our system obtains temperature measurements through a self-calibrating grid of sensors and uses area constrained heat-generating circuits in order to generate spatial and temporal temperature gradients. The generated temperature variations are then used to learn the free parameters of the system's thermal model. The system thus acquires an understanding of its own thermal characteristics. We implemented an FPGA system containing a net of 144 temperature sensors on a Xilinx Virtex-6 LX240T FPGA that is aware of its thermal model. Finally, we show that the temperature predictions vary less than 0.72 degree C on average compared to the measured temperature distributions at run-time.


MoSGrid: Progress of Workflow driven Chemical Simulations

G. Birkenheuer, D. Blunk, S. Breuers, A. Brinkmann, G. Fels, S. Gesing, R. Grunzke, S. Herres-Pawlis, O. Kohlbacher, J. Krüger, U. Lang, L. Packschies, R. Müller-Pfefferkorn, P. Schäfer, J. Schuster, T. Steinke, K. Warzecha, M. Wewior, in: Proc. of Grid Workflow Workshop (GWW), 2011


Parallel Monte-Carlo Tree Search for HPC Systems

T. Graf, U. Lorenz, M. Platzner, L. Schaefers, in: Proc. European Conf. on Parallel Processing (Euro-Par), Springer, 2011

DOI


Performance Estimation Framework for Automated Exploration of CPU-Accelerator Architectures

T. Kenter, M. Platzner, C. Plessl, M. Kauschke, in: Proc. Int. Symp. on Field-Programmable Gate Arrays (FPGA), ACM, 2011, pp. 177-180

DOI


Reliable and Randomized Data Distribution Strategies for Large Scale Storage Systems

A. Miranda, S. Effert, Y. Kang, E. Miller, A. Brinkmann, T. Cortes, in: Proc. Int. Conf. on High Performance Computing (HIPC), IEEE Computer Society, 2011, pp. 1-10

DOI


Request Load Balancing for Highly Skewed Traffic in P2P Networks

A. Brinkmann, Y. Gao, M. Korzeniowski, D. Meister, in: Proc. IEEE Int. Conf. on Networking, Architecture and Storage (NAS), IEEE, 2011, pp. 53-62

DOI


Rule Based Mapping of Virtual Machines in Clouds

C. Kleineweber, A. Keller, O. Niehörster, A. Brinkmann, in: Proc. Int. Conf. on Parallel, Distributed and Network-Based Computing (PDP), 2011

Infrastructure as a Service providers use virtualization to abstract their hardware and to create a dynamic data center. Virtualization enables the consolidation of virtual machines as well as the migration of them to other hosts during runtime. Each provider has its own strategy to efficiently operate a data center. We present a rule based mapping algorithm for VMs, which is able to automatically adapt the mapping between VMs and physical hosts. It offers an interface where policies can be defined and combined in a generic way. The algorithm performs the initial mapping at request time as well as a remapping during runtime. It deals with policy and infrastructure changes. We extended the open source IaaS solution Eucalyptus and we evaluated it with typical policies: maximizing the compute performance and VM locality to achieve a high performance and minimizing energy consumption. The evaluation was done on state-of-the-art servers in our own data center and by simulations using a workload of the Parallel Workload Archive. The results show that our algorithm performs well in dynamic data centers environments.


Transformation of scientific algorithms to parallel computing code: subdomain support in a MPI-multi-GPU backend

B. Meyer, C. Plessl, J. Förstner, in: Symp. on Application Accelerators in High Performance Computing (SAAHPC), IEEE Computer Society, 2011, pp. 60-63

DOI


Virtualized HPC: a contradiction in terms?

G. Birkenheuer, A. Brinkmann, J. Kaiser, A. Keller, M. Keller, C. Kleineweber, C. Konersmann, O. Niehörster, T. Schäfer, J. Simon, M. Wilhelm, Software: Practice and Experience (2011)

System virtualization has become the enabling technology to manage the increasing number of different applications inside data centers. The abstraction from the underlying hardware and the provision of multiple virtual machines (VM) on a single physical server have led to a consolidation and more efficient usage of physical servers. The abstraction from the hardware also eases the provision of applications on different data centers, as applied in several cloud computing environments. In this case, the application need not adapt to the environment of the cloud computing provider, but can travel around with its own VM image, including its own operating system and libraries. System virtualization and cloud computing could also be very attractive in the context of high‐performance computing (HPC). Today, HPC centers have to cope with both, the management of the infrastructure and also the applications. Virtualization technology would enable these centers to focus on the infrastructure, while the users, collaborating inside their virtual organizations (VOs), would be able to provide the software. Nevertheless, there seems to be a contradiction between HPC and cloud computing, as there are very few successful approaches to virtualize HPC centers. This work discusses the underlying reasons, including the management and performance, and presents solutions to overcome the contradiction, including a set of new libraries. The viability of the presented approach is shown based on evaluating a selected parallel, scientific application in a virtualized HPC environment.


2010

An Open Source Circuit Library with Benchmarking Facilities

M. Grad, C. Plessl, in: Proc. Int. Conf. on Engineering of Reconfigurable Systems and Algorithms (ERSA), CSREA Press, 2010, pp. 144-150


Balls into Bins with Related Random Choices

P. Berenbrink, A. Brinkmann, T. Friedetzky, L. Nagel, in: Proc. Int. Symp. on Parallelism in Algorithms and Architectures (SPAA), ACM, 2010, pp. 100-105

DOI


Balls into Non-uniform Bins

P. Berenbrink, A. Brinkmann, T. Friedetzky, L. Nagel, in: Proc. Int. Symp. on Parallel and Distributed Processing (IPDPS), IEEE, 2010, pp. 1-10

DOI


Configurable Processor Architectures: History and Trends

D. Andrews, C. Plessl, in: Proc. Int. Conf. on Engineering of Reconfigurable Systems and Algorithms (ERSA), CSREA Press, 2010, pp. 165


dedupv1: Improving Deduplication Throughput using Solid State Drives (SSD)

D. Meister, A. Brinkmann, in: Proc. Symp. on Mass Storage Systems and Technologies (MSST), IEEE Computer Society, 2010, pp. 1-6

DOI


Enforcing SLAs in Scientific Clouds

O. Niehörster, A. Brinkmann, G. Fels, J. Krüger, J. Simon, in: Proc. Int. Conf. on Cluster Computing (CLUSTER), IEEE, 2010, pp. 178-187

DOI


Grid-Workflows in Molecular Science

G. Birkenheuer, S. Breuers, A. Brinkmann, D. Blunk, G. Fels, S. Gesing, S. Herres-Pawlis, O. Kohlbacher, J. Krüger, L. Packschies, in: Proc. of Grid Workflow Workshop (GWW), Gesellschaft für Informatik (GI), 2010, pp. 177-184


hashFS: Applying Hashing to Optimized File Systems for Small File Reads

P.H. Lensing, D. Meister, A. Brinkmann, in: Proc. Int. Worksh. on Storage Network Architecture and Parallel I/Os (SNAPI), IEEE, 2010, pp. 33-42

DOI


Non-intrusive Virtualization Management Using libvirt

M. Bolte, M. Sievers, G. Birkenheuer, O. Niehörster, A. Brinkmann, in: Proc. Design, Automation and Test in Europe Conf. (DATE), EDA Consortium, 2010


Open Source Middleware for Networked Embedded Systems towards Future Internet of Things

N. R. Prasad, M. Eisenhauer, M. Ahlsén, A. Badii, A. Brinkmann, K. Marius Hansen, P. Rosengren, in: Vision and Challenges for Realising the Internet of Things, European Commission, 2010, pp. 153-163


Performance Estimation for the Exploration of CPU-Accelerator Architectures

T. Kenter, M. Platzner, C. Plessl, M. Kauschke, in: Proc. Workshop on Architectural Research Prototyping (WARP), International Symposium on Computer Architecture (ISCA), 2010


Proc. Int. Conf. on Engineering of Reconfigurable Systems and Algorithms (ERSA)

T.P. Plaks, D. Andrews, R. DeMara, H. Lam, J. Lee, C. Plessl, G. Stitt, CSREA Press, 2010


Pruning the Design Space for Just-In-Time Processor Customization

M. Grad, C. Plessl, in: Proc. Int. Conf. on ReConFigurable Computing and FPGAs (ReConFig), IEEE Computer Society, 2010, pp. 67-72

DOI


Reconfigurable Nodes for Future Networks

A. Keller, B. Plattner, E. Lübbers, M. Platzner, C. Plessl, in: Proc. IEEE Globecom Workshop on Network of the Future (FutureNet), IEEE, 2010, pp. 372-376

DOI


Reliability Analysis of Declustered-Parity RAID 6 with Disk Scrubbing and Considering Irrecoverable Read Errors

Y. Gao, D. Meister, A. Brinkmann, in: Proc. IEEE Int. Conf. on Networking, Architecture and Storage (NAS), IEEE, 2010, pp. 126-134

DOI


Risikomanagement für verteilte Umgebungen

A. Brinkmann, D. Battré, G. Birkenheuer, O. Kao, K. Voß, ForschungsForum Paderborn (2010), 13(13)


Risk Aware Overbooking for Commercial Grids

G. Birkenheuer, A. Brinkmann, H. Karl, in: Job Scheduling Strategies for Parallel Processing - 15th International Workshop, JSSPP 2010, Atlanta, GA, USA, April 23, 2010, Revised Selected Papers, 2010, pp. 51-76

DOI


Rupeas: Ruby Powered Event Analysis DSL

M. Woehrle, C. Plessl, L. Thiele, in: Proc. Int. Conf. Networked Sensing Systems (INSS), IEEE, 2010, pp. 245-248

DOI


SkewCCC+: A Heterogeneous Distributed Hash Table

M. Bienkowski, A. Brinkmann, M. Klonowski, M. Korzeniowski, in: Proceedings of the 14th International Conference On Principles Of Distributed Systems (Opodis), Springer, 2010

DOI


The MoSGrid Gaussian Portlet - Technologies for the Implementation of Portlets for Molecular Simulations

M. Wewior, L. Packschies, D. Blunk, D. Wickeroth, K. Warzecha, S. Herres-Pawlis, S. Gesing, S. Breuers, J. Krüger, G. Birkenheuer, U. Lang, in: Proc. Int. Workshop on Scientific Gateways (IWSG), Consorzio COMETA, 2010, pp. 39-43


Towards Adaptive Networking for Embedded Devices based on Reconfigurable Hardware

E. Lübbers, M. Platzner, C. Plessl, A. Keller, B. Plattner, in: Proc. Int. Conf. on Engineering of Reconfigurable Systems and Algorithms (ERSA), CSREA Press, 2010, pp. 225-231


Using Shared Library Interposing for Transparent Acceleration in Systems with Heterogeneous Hardware Accelerators

T. Beisel, M. Niekamp, C. Plessl, in: Proc. Int. Conf. on Application-Specific Systems, Architectures, and Processors (ASAP), IEEE Computer Society, 2010, pp. 65-72

DOI


Workflow Interoperability in a Grid Portal for Molecular Simulations

S. Gesing, I. Marton, G. Birkenheuer, B. Schuller, R. Grunzke, J. Krüger, S. Breuers, D. Blunk, G. Fels, L. Packschies, A. Brinkmann, O. Kohlbacher, M. Kozlovszky, in: Proc. Int. Workshop on Scientific Gateways (IWSG), Consorzio COMETA, 2010, pp. 44-48


2009

A Microdriver Architecture for Error Correcting Codes inside the Linux Kernel

A. Brinkmann, D. Eschweiler, Journal of Supercomputing (2009), pp. 35:1-35:10

DOI


An Accelerator for k-th Nearest Neighbor Thinning Based on the IMORC Infrastructure

T. Schumacher, C. Plessl, M. Platzner, in: Proc. Int. Conf. on Field Programmable Logic and Applications (FPL), IEEE, 2009, pp. 338-344


An Orchestration as a Service Infrastructure using Grid Technologies and WS-BPEL

A. Höing, G. Scherp, S. Gudenkauf, D. Meister, A. Brinkmann, in: Proc. Int. Conf. on Service Oriented Computing (ICSOC), Springer, 2009, pp. 301-315

DOI


Communication Performance Characterization for Reconfigurable Accelerator Design on the XD1000

T. Schumacher, T. Süß, C. Plessl, M. Platzner, in: Proc. Int. Conf. on ReConFigurable Computing and FPGAs (ReConFig), IEEE Computer Society, 2009, pp. 119-124

DOI


Connecting Communities on the Meta-Scheduling Level: The DGSI Approach!

G. Birkenheuer, A. Carlson, A. Fölling, M. Högqvist, A. Hoheisel, A. Papaspyrou, K. Rieger, B. Schott, W. Ziegler, in: Proc. Cracow Grid Workshop (CGW), 2009, pp. 96-103


EvoCaches: Application-specific Adaptation of Cache Mapping

P. Kaufmann, C. Plessl, M. Platzner, in: Proc. NASA/ESA Conference on Adaptive Hardware and Systems (AHS), IEEE Computer Society, 2009, pp. 11-18

In this work we present EvoCache, a novel approach for implementing application-specific caches. The key innovation of EvoCache is to make the function that maps memory addresses from the CPU address space to cache indices programmable. We support arbitrary Boolean mapping functions that are implemented within a small reconfigurable logic fabric. For finding suitable cache mapping functions we rely on techniques from the evolvable hardware domain and utilize an evolutionary optimization procedure. We evaluate the use of EvoCache in an embedded processor for two specific applications (JPEG and BZIP2 compression) with respect to execution time, cache miss rate and energy consumption. We show that the evolvable hardware approach for optimizing the cache functions not only significantly improves the cache performance for the training data used during optimization, but that the evolved mapping functions generalize very well. Compared to a conventional cache architecture, EvoCache applied to test data achieves a reduction in execution time of up to 14.31% for JPEG (10.98% for BZIP2), and in energy consumption by 16.43% for JPEG (10.70% for BZIP2). We also discuss the integration of EvoCache into the operating system and show that the area and delay overheads introduced by EvoCache are acceptable.


IMORC: Application Mapping, Monitoring and Optimization for High-Performance Reconfigurable Computing

T. Schumacher, C. Plessl, M. Platzner, in: Proc. Int. Symp. on Field-Programmable Custom Computing Machines (FCCM), IEEE Computer Society, 2009, pp. 275-278

Mapping applications that consist of a collection of cores to FPGA accelerators and optimizing their performance is a challenging task in high performance reconfigurable computing. We present IMORC, an architectural template and highly versatile on-chip interconnect. IMORC links provide asynchronous FIFOs and bitwidth conversion which allows for flexibly composing accelerators from cores running at full speed within their own clock domains, thus facilitating the re-use of cores and portability. Further, IMORC inserts performance counters for monitoring runtime data. In this paper, we first introduce the IMORC architectural template and the on-chip interconnect, and then demonstrate IMORC on the example of accelerating the k-th nearest neighbor thinning problem on an XD1000 reconfigurable computing system. Using IMORC's monitoring infrastructure, we gain insights into the data-dependent behavior of the application which, in turn, allow for optimizing the accelerator.


Multi-Level Comparison of Data Deduplication in a Backup Scenario

D. Meister, A. Brinkmann, in: Proc. of the Israeli Experimental Systems Conference (SYSTOR), ACM, 2009, pp. 8:1-8:12

DOI


PermaDAQ: A Scientific Instrument for Precision Sensing and Data Recovery in Environmental Extremes

J. Beutel, S. Gruber, A. Hasler, R. Lim, A. Meier, C. Plessl, I. Talzi, L. Thiele, C. Tschudin, M. Woehrle, M. Yuecel, in: Proc. Int. Conf. on Information Processing in Sensor Networks (IPSN), IEEE Computer Society, 2009, pp. 265-276


Providing Scientific Software as a Service in Consideration of Service Level Agreements

O. Niehörster, G. Birkenheuer, A. Brinkmann, D. Blunk, B. Elsässer, S. Herres-Pawlis, J. Krüger, J. Niehörster, L. Packschies, G. Fels, in: Proc. Cracow Grid Workshop (CGW), 2009, pp. 55-63


Rupeas: Ruby Powered Event Analysis DSL

M. Woehrle, C. Plessl, L. Thiele, 2009

Wireless Sensor Networks (WSNs) are unique embedded computation systems for distributed sensing of a dispersed phenomenon. While being a strongly concurrent distributed system, its embedded aspects with severe resource limitations and the wireless communication requires a fusion of technologies and methodologies from very different fields. As WSNs are deployed in remote locations for long-term unattended operation, assurance of correct functioning of the system is of prime concern. Thus, the design and development of WSNs requires specialized tools to allow for testing and debugging the system. To this end, we present a framework for analyzing and checking WSNs based on collected events during system operation. It allows for abstracting from the event trace by means of behavioral queries and uses assertions for checking the accordance of an execution to its specification. The framework is independent from WSN test platforms, applications and logging semantics and thus generally applicable for analyzing event logs of WSN test executions.


The Gain of Overbooking

G. Birkenheuer, A. Brinkmann, H. Karl, in: Job Scheduling Strategies for Parallel Processing, 14th International Workshop, JSSPP 2009, Rome, Italy, May 29, 2009. Revised Papers, 2009, pp. 80-100

DOI


Woolcano: An Architecture and Tool Flow for Dynamic Instruction Set Extension on Xilinx Virtex-4 FX

M. Grad, C. Plessl, in: Proc. Int. Conf. on Engineering of Reconfigurable Systems and Algorithms (ERSA), CSREA Press, 2009, pp. 319-322

In this paper, we introduce the Woolcano reconfigurable processor architecture. The architecture is based on the Xilinx Virtex-4 FX FPGA and leverages the Auxiliary Processing Unit (APU) as well as the partial reconfiguration capabilities to provide dynamically reconfigurable custom instructions. We also present a hardware tool flow that automatically translates software functions into custom instructions and a software tool flow that creates binaries using these instructions. While previous research on processors with reconfigurable functional units has been performed predominantly with simulation, the Woolcano architecture allows for exploring dynamic instruction set extension with commercially available hardware. Finally, we present a case study demonstrating a custom floating-point instruction generated with our approach, which achieves a 40x speedup over software-emulated floating-point operations and a 21% speedup over the Xilinx hardware floating-point unit.


2008

A Hardware Accelerator for k-th Nearest Neighbor Thinning

T. Schumacher, R. Meiche, P. Kaufmann, E. Lübbers, C. Plessl, M. Platzner, in: Proc. Int. Conf. on Engineering of Reconfigurable Systems and Algorithms (ERSA), CSREA Press, 2008, pp. 245-251


A method for OSEM PET reconstruction on parallel architectures using STIR

T. Beisel, S. Lietsch, K. Thielemans, in: IEEE Nuclear Science Symposium Conference Record (NSS), IEEE, 2008, pp. 4161-4168

DOI


Applying Risk Management to Support SLA Provisioning

D. Battre, G. Birkenheuer, M. Hovestadt, O. Kao, K. Voss, in: Proc. Cracow Grid Workshop (CGW), 2008


Computational Steering Of Interactive Material Flow Simulations

S. Lietsch, H. Zabel, C. Laroque, in: Proc. ASME Computers and Information in Engineering Conference (CIE), ASME, 2008, pp. 1493-1502

DOI


Computational Steering verteilter, interaktiver Materialflusssimulationen

C. Laroque, S. Lietsch, H. Zabel, in: Augmented & Virtual Reality in der Produktentstehung, Heinz Nixdorf Institut, 2008, pp. 221-239


CUDA-based, parallel JPEG Compression for Remote Rendering

S. Lietsch, P. Hermann Lensing, in: Proc. Int. Symp. on Image/Video Communications over fixed and mobile networks (ISVC), IEEE, 2008


Data Replication in P2P Environments

A. Brinkmann, S. Effert, in: Proc. Int. Symp. on Parallelism in Algorithms and Architectures (SPAA), ACM, 2008, pp. 191-193

DOI


Degree 3 Suffices: A Large-Scale Overlay for P2P Networks

M. Bienkowski, A. Brinkmann, M. Korzeniowski, in: Proc. Int. Conf. on Principles Of DIstributed Systems (OPODIS), Springer, 2008, pp. 184-196

DOI


Employing WS-BPEL Design Patterns for Grid Service Orchestration using a Standard WS-BPEL Engine and a Grid Middleware

A. Brinkmann, S. Gudenkauf, W. Hasselbring, A. Höing, H. Karl, O. Kao, H. Nitsche, G. Scherp, in: Proc. Cracow Grid Workshop (CGW), 2008, pp. 103-110


Enhancing SLA Provisioning by Utilizing Profit-Oriented Fault Tolerance

D. Battré, M. Hovestadt, O. Kao, A. Keller, K. Voss, in: Proc. Int. Conf. on Parallel and Distributed Computing and Systems (PDCS), 2008, pp. 212-218


EvAnT: Analysis and Checking of event traces for Wireless Sensor Networks

M. Woehrle, C. Plessl, R. Lim, J. Beutel, L. Thiele, in: IEEE Int. Conf. on Sensor Networks, Ubiquitous, and Trustworthy Computing (SUTC), IEEE Computer Society, 2008, pp. 201-208

DOI


Germany, Belgium, France, and Back Again: Job Migration using Globus

D. Battré, M. Hovestadt, O. Kao, A. Keller, K. Voss, in: Proc. Int. Conf. on Grid Computing and Applications (GCA), 2008


GPU-Supported Image Compression for Remote Visualization - Realization and Benchmarking

S. Lietsch, P. Hermann Lensing, in: Proc. Int. Symp. on Visual Computing (ISVC), Springer, 2008, pp. 658-668

DOI


Guarantee and Penalty Clauses for Service Level Agreements

D. Battré, G. Birkenheuer, V. Deora, M. Hovestadt, O. Rana, O. Wäldrich, in: Proc. Cracow Grid Workshop (CGW), 2008, pp. 213-220


Guiding exploration by combining individual learning and imitation in societies of autonomous robots

W. Richert, F. Klompmaker, O. Niehörster, in: Proc. IFIP Conf. on Biologically Inspired Cooperative Computing (BICC), Springer, 2008, pp. 233-244

DOI


IMORC: An infrastructure for performance monitoring and optimization of reconfigurable computers

T. Schumacher, C. Plessl, M. Platzner, in: Many-core and Reconfigurable Supercomputing Conference (MRSC), 2008


Implementation of Virtual Execution Environments for improving SLA-compliant Job Migration in Grids

D. Battré, M. Hovestadt, O. Kao, A. Keller, K. Voss, in: Proc. Int. Workshop on Scheduling and Resource Management for Parallel and Distributed Systems, 2008

Abstract: Commercial Grid users demand for contractually fixed QoS levels. Service Level Agreements (SLAs) are powerful instruments for describing such contracts. SLA-aware resource management is the foundation for realizing SLA contracts within the Grid. OpenCCS is such an SLA-aware RMS, using transparent checkpointing to cope with resource outages. It generates a compatibility profile for each checkpoint dataset, so that the job can be resumed even on resources within the Grid. However, only a small number of Grid resources comply to such a profile. This paper describes the concept of virtual execution environments and how they increase the number of potential migration targets.The paper also describes how these virtual execution environments have been implemented within the OpenCCS resource management system.


Implementing WS-Agreement in a Globus Toolkit 4.0 Environment

D. Battré, O. Kao, K. Voss, in: Proc. Usage of Service Level Agreements in Grids Workshop held in conjunction with International Conference on Grid Computing, Springer, 2008, pp. 409-418

DOI


Liste im Research Information System öffnen

Die Universität der Informationsgesellschaft