Parallelism in MM5
What is meant by “parallel”?
- Increase computational and memory resources available for larger, faster runs by having more than one computer work on the problem
Isn’t MM5 already parallel?
- Yes, the model has been able to run shared-memory parallel since MM4 using Cray Microtasking directives
- More recently, standardized OpenMP directives have been added
What is DM-parallelism? Why?
- Processors store part of model domain in local memory, not shared with other processors; work together on a problem by exchanges messages over a network
- “Scalable” because it eliminates bottlenecks on shared resources such as bus or memory
- Cost effective since systems can be built from “commodity” components
You already have the DM-parallel version of MM5