Hopper Scaling Incentive Program
August 30, 2011 by Francesca Verdier (0 Comments)
For projects that haven't yet scaled their codes to 683 or more nodes (which is the level at which a job is considered "big" on hopper) NERSC is offering scaling incentives, mostly focused on the use of OpenMP. For some codes, adding OpenMP directives will allow you to scale up and run bigger science problems.
For users accepted in the Scaling Incentive Program:
- First, you'll need to understand where your code spends the most time. If you don't have a recent code timing profile, a NERSC consultant can assist you in making one.
- Second, you'll need to develop a plan for improving the scaling characteristics of your code. While NERSC cannot provide a programmer to modify your code, we can discuss options with you. Scaling improvements can be accomplished in a number of ways, including use of OpenMP or threads, or better use of MPI. For this program we would like to encourage use of OpenMP, which, in some cases, can be added to a code incrementally, and can result in not only better performance but also improved memory usage on the node.
- Third, any runs you make using 683 or more nodes will be reimbursed - up to 1 million MPP hours per user with a limit of 2 million hours per repository.
To request enrollment: send email to email@example.com with the subject line "Scaling Incentive Program".