Dhabaleswar Panda (pictured), professor and university distinguished scholar at The Ohio State University, has worked on many things, having developed the MVAPICH (High Performance MPI and PGAS over InfiniBand, Omni-Path, iWARP and RoCE) libraries with his research group, he is a distinguished scholar of computer science at the Ohio State University.
“Ohio State University is one of the largest single campuses in the U.S., one of the top three, top four,” he said. “We have 65,000 students, and especially within computer science where I am located, high-performance computing is a very big focus.”
Panda spoke with theCUBE industry analysts Paul Gillin and David Nicholson at SC22, during an exclusive broadcast on theCUBE, SiliconANGLE Media’s livestreaming studio. They discussed node-based computing and recent developments. (* Disclosure below.)
Development of scalable node clusters
The high-performance computing field is moving to integrate artificial intelligence and big data in an effort to provide faster computation and data interpretation. It is critical to manage scalability of core counts in node computing, according to Panda.
“Where we look at the vision for the next 20 years, we want to design this MPI library so that not only HPC, but also all other workloads can take advantage of it,” Panda stated.
The challenge is that while people use these frameworks, models are continuously becoming larger. This necessitates a very fast turnaround time, Panda added.
“So how do you train faster? How do you influence faster? So this is where HPC comes in,” he stated. “We have actually done that kind of a tight coupling, and that helps the research to really take advantage of HPC. ”
Here’s the complete video interview, part of SiliconANGLE’s and theCUBE’s coverage of the SC22 event:
(* Disclosure: This is an unsponsored editorial segment. However, theCUBE is a paid media partner for SC22. Neither Dell Technologies Inc., the main sponsor for theCUBE’s event coverage, nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)