Since entering service in May 2015, Comet has established itself as one of the most widely used supercomputers in the NSF's eXtreme Science and Engineering Discovery Environment (XSEDE) programme, which connects researchers to an advanced collection of integrated digital resources and services.
"Comet has served researchers as a key resource among a wide range of science domains and is an important resource for the analysis of data from large scientific instruments that led to recent landmark discoveries by scientists at the Laser Interferometer Gravitational Wave Observatory (LIGO), the IceCube Observatory, and XENON Collaboration", stated SDSC Director Michael Norman.
While Comet is capable of an overall peak performance of two petaflops - or two quadrillion calculations per second - its design, allocation, and operational policies are focused on maximizing scientific productivity by lowering barriers to accessing HPC resources through science gateways, rapid access accounts, and job scheduling optimizations targeted at modest scale computing jobs.
Since Comet entered production in 2015, it has served more than 1,700 unique PIs at some 350 institutions. Today, there are more than 1,300 active allocations covering all 50 U.S. states.
"Advanced but user-friendly resources such as the Comet supercomputer underscore a vital need for systems that serve a broad range of research, with a focus on researchers who have modest-scale computational needs", stated SDSC Deputy Director Shawn Strande. "From the outset, Comet was configured the serve the long tail of science, or the idea that the large number of modest-sized computationally-based research projects represents, in aggregate, a tremendous amount of research and scientific advances."
SDSC recently doubled the number of graphic processing units (GPUs) on Comet in direct response to growing demand for GPU computing among a wide range of research domains. The expansion, made under an earlier supplemental NSF award valued at about $900,000, provided for the addition of 36 GPU nodes, each with four NVIDIA P100s, doubling the number of GPUs from the previous 144 to 288.
That award makes Comet the largest GPU resource available through the XSEDE programme. Applications include but are not limited to GPU-memory management systems such as VAST, analysis of data from large scientific instruments, and molecular dynamics software packages such as AMBER, LAMMPS, and BEAST - the latter used extensively by SDSC's Cyberinfrastructure for Phylogenetic Research (CIPRES) science gateway, which receives the majority of its computing resources from Comet.
The NSF award number is 1341698.