Refactoring and Optimizing the Community Atmosphere Model (CAM) on the New Sunway Many-Core Supercomputer
SessionAccelerating Science
Session ChairMichael Bader
Authors
Event Type
Paper
Applications
Heterogeneous Systems
Intermediate
Performance
Scientific Computing
TimeThursday, November 17th4pm - 4:30pm
Location255-EF
DescriptionThis paper reports our efforts in refactoring and optimizing the Community Atmosphere Model(CAM) on the Sunway TaihuLight supercomputer, which uses many-core processors that consist of management processing elements (MPEs) and clusters of computing processing elements (CPEs). To map the large code base of CAM to the millions of cores on the Sunway system, we take OpenACC-based refactoring as the major approach, and apply source-to-source translator tools to exploit the most suitable parallelism for the CPE cluster, and to fit the intermediate variable into the limited on-chip fast buffer. For individual kernels, when comparing the original ported version using only MPEs and the refactored version using both the MPE and CPE clusters, we achieve up to 22x speedup for the compute-intensive kernels. For the 25km resolution CAM global model, we manage to scale to 24,000 MPEs and 1,536,000 CPEs, and we achieve a simulation speed of 2.81 model years per day.
Download PDF
Paper provided by the IEEE Computer SocietyPaper also available from the ACM Digital Library
Authors
