Advanced Computing, Mathematics and Data
Staff Awards & Honors

March 2015

HPC Team Tackles Parallel Programming Model Performance at HiPC

As part of the technical program during the 21st annual IEEE International Conference on High Performance Computing, or HiPC 2014, held last December in Goa, India, Abhinav Vishnu, a research scientist with PNNL’s High Performance Computing group, presented research on designing scalable communication runtime for extreme scale (ComEx) as highlighted in the paper “On the Use of MPI as a PGAS Runtime.” Vishnu, working with other HPC group partners, Jeff Daily (the paper’s lead author), Bruce Palmer, and Darren Kerbyson, as well as Hubertus van Dam, of EMSL’s Molecular Science Computing capability, showed how their work deals with a problem facing many PGAS (partitioned global address space) models: how to scale PGAS applications, such as NWChem, when a supercomputer comes online.

Daily, Vishnu, van Dam, Palmer, and Kerbyson (from left) presented their work with the PGAS model at HiPC 2014. Enlarge Image.

As part of his presentation, Vishnu showed how the team proposed several solutions to handle this scaling problem and demonstrated that their approach is performance portable for high-end U.S. Department of Energy computing systems in the field.

HiPC is an international HPC systems forum focused on all aspects of their scientific, engineering, and commercial applications and showcases current research gleaned from computing experts worldwide. In 2015, HiPC will be held in Bengaluru, India.

Reference:
Daily J, A Vishnu, H van Dam, B Palmer, and D Kerbyson. 2014. “On the Use of MPI as a PGAS Runtime.” Presented at: 21st Annual IEEE International Conference on High Performance Computing (HiPC 2014). December 17-20, 2014, Goa, India.