MPI/OMP Hybrid Multi-threading
EFDC+ has been optimized to apply a domain decomposition approach based on Intel’s Message Passing Interface (MPI). With this feature, EFDC+ can run across large cluster systems as well as multi-core desktop computers with greater speed than ever before. EFDC+ with OMP typically produced run times up to 4 times faster on a six-core processor than the conventional single-threaded EFDC model. Testing on large domain models (500K cells) demonstrates that this new hybrid MPI/OMP approach achieves a speedup of nearly 17 with 32 compute cores. Using a cluster conﬁgured on Amazon Web Services (AWS), a strong scaling study was conducted, which resulted in a speedup of nearly 25 while using 96 compute cores.
Overall, the domain decomposition approach using MPI provides the ability to run across a cluster and oﬀers substantially improved performance compared to previous versions of EFDC+. To learn about the performance gains achievable with this new approach, as well as guidance on configuring models read, our white paper.