The Path to Exascale Computing
Speaker: Alessandro Morari and Abdullah Kay, IBM T.J. Watson Research Center
Location: 60 Fifth Avenue 150
Date: October 27, 2017, 11 a.m.
Host: Mohamed Zahran
For decades, Supercomputers have been at the frontier of computing technology and opened new ways for scientific discovery. After almost a decade since reaching Petascale, the challenge on the horizon is to reach Exascale computing. The creation of an Exascale system will have major impacts in key areas including national security, medicine, finance, and many other fields. In this talk, we will go over some of the basics about Supercomputing (High Performance Computing), talk about the IBM CORAL System and then focus on Exascale computing. We also talk about the current software/hardware trends and give an overview of the challenges that the industry is facing to build an Exascale system.
Dr. Alessandro Morari is the manager of the System Software team in the Data Centric Systems department at IBM Research. This team performs research and development of software at the intersection between traditional high performance computing and next generation data-centric systems. His main research interests include: Data-centric computing, High Performance Computing, Parallel Computing, Performance Analysis, and Machine Learning. Currently, he is investigating the use of High Performance Computing techniques to accelerate the software stack of large scale workloads, with a particular focus on data intensive and machine learning algorithms. He and his team have also been involved in the development of software for the CORAL Supercomputer commissioned by the US Department of Energy.
Dr. Abdullah Kayi is a Research Staff Member in the Data-Centric Solutions department at IBM T.J. Watson Research Center. His research interests include high-performance computing, parallel programming languages, runtimes, and computer architecture. He is currently working on software/hardware co-design of future high-performance systems on the path to Exascale computing. He has made significant contributions to the Partitioned Global Address Space (PGAS) languages and he was one of the developers in the Unified Parallel C (UPC) consortium. Prior to joining IBM, he was a Research Scientist at Intel Labs and also worked on various other groups including the Intel Architecture group.
Refreshments will be offered starting 15 minutes prior to the scheduled start of the talk.