Any long running (total CPU run time
> 2hrs) jobs on dev nodes will be killed automatically without advance notice.
The HPCC has several development nodes that are available for users to compile their code and do short runs (less than 2 hours) to estimate run-time and memory usage.
These development nodes run the latest operating system and have similar configurations and environment setups as the compute nodes of the same clusters. Please use these development nodes to compile your program and test the work flow of your job script. For running long-time or large-resource computations, please submit jobs to use compute nodes.
Code compiled on older development nodes (dev-intel14 and dev-intel14-k20) may have errors when running on the latest clusters due to an outdated instruction set. To resolve this, compile your code on a newer development node.
ssh to the development nodes after
connecting to the gateway via SSH.
To access a certain development node, for example dev-amd20, please
ssh dev-amd20 from the gateway. Users may also directly
connect to development nodes by setting up SSH Tunneling
Alternatively, they may be accessed through the "Development Nodes" tab on OnDemand.
Nodes with -k80 or -v100 suffixes have GPU cards required by GPU-enabled software, but may be used for any software. Note there is not a development node containing the AMD20 A100 GPUs.
|dev-amd20||128||960GB||AMD EPYC 7H12 64-Core Processor @ 2.6GHz|
|dev-amd20-v100||48||187GB||Intel(R) Xeon(R) Platinum 8260 CPU @ 2.40GHz and 4 Tesla V100S|
|dev-intel18||40||377GB||Two 2.4Ghz 20-core Intel Xeon Gold 6148 CPU (40 cores total)|
|dev-intel16||28||128GB||Two 2.4Ghz 14-core Intel Xeon E5-2680v4 (28 cores total)|
|dev-intel16-k80||28||256GB||Intel16 node with 4 Nvidia Tesla K80 GPUs|
Once your program is compiled and job script is tested, users can submit it to the SLURM queue by specifying various constraints such as job duration, memory usage, number of CPUs, software license reservations and so on.