Step 3: Apply for Resources
If you have identified a suitable project type in Step 2: Find Right Project Type, you can find on this page the right guide how to apply for your project type and further information.
| Detailed Description Templates | |||
|---|---|---|---|
| Project Type | Guides to Apply | Already Scientifically Reviewed | Regular |
| NHR-large | Guide to apply for NHR-large or NHR-normal | not possible | DOCX and TeX, use new whitelisted variant |
| NHR-normal | DOCX and TeX, use new whitelisted variant | ||
| NHR-starter | Guide to apply for NHR-starter | not required | |
| small | Guide to apply for a small project | not required | |
| fpga | |||
| test | Guide to apply for a test project | not required | |
The main NHR page for templates and computing time can be found at https://go-nhr.de/computing-time.
Hints and Examples
Please don't forget to consider the memory usage in your resource request/justification: In the accounting of used compute time, the memory reserved, i.e., the memory allocation specified in the job request, counts towards your resource usage of CPU-core-hours. For example, if you proportionally allocate more memory of the node than CPU cores, then the higher of the two determines the effective number of CPU-core-h used by the job. This is done so that the memory allocation in jobs is accounted in a fair way.
Example: allocated 4 CPU cores and 120 GB of main memory of a Otus normal node (730 GB useable main memory), e.g. sbatch --mem 120000 -N 1 --ntask-per-node 4 --cpus-per-task 1, would “cost” you for a one-hour job 31.56 CPU-core-hours (1 hour times 192 cores per node times 120 GB/730 GB) instead of 4 CPU-core-hours.
- Please consider using the specialized resource justification table, which is designed in terms of the common units ns/day.
Also workloads with unfavorable parallel scaling, e.g. not scaling beyond a few CPU cores or single GPUs, can be granted computation time if the HPC systems
offered by NHR are required to achieve progress for the given scientific problem that would not be possible with other HPC systems. The same holds for embarrassingly parallel workloads (high-throughput computing, e.g., parameter studies ) where the total computational need requires a large
HPC system offered by NHR. In these cases, no scalability benchmarks are needed.
If you have applied for resources you can proceed to the next step in the application process.