Performance Advantages of a Data Centric Computer Architecture

Speaker: Ken Jacobsen, CEO of EMU Technology
December 15, 2016

Presentation

Abstract

There is growing evidence that traditional computer architectures do not handle applications well where the data is not cache-friendly, such as sparse matrix operation, data analytics, machine learning and graph analysis. This is partly due to memory access characteristics, and partly how accesses to remote memory are handled for many core machine implementations. This talk will introduce a new highly-scalable PGAS architecture that can scale to very large sizes using a shared memory programming model, and does so relatively invisibly to the programmer. In this architecture, the program thread migrates to the data, and data never moves, which can result in dramatic performance improvements.

Biography

Ken Jacobsen has more than 35 years of experience building and leading highly technical organizations. Most recently he was the CEO at GreenWave Systems Inc., a company designing an exascale capable computer architecture. Previously, he lead all technical efforts at the Morse Project, creating smart devices for the Internet of Things, and was the CEO of Foveal Recall Inc., which created specialized software tools for image object characterization and recognition. He has also held the position as EVP of Informatics at Incyte Genomics, and VP of Applications at SGI. Ken holds an undergraduate degree from the California Institute of Technology and a Ph.D. from the University of California at Berkeley.

Seminar Participation

Attending In Person

For select seminars, you may wish to attend in-person if you are in the vicinity of NASA Ames Research Center. These AMS seminars are held in Building N258 in Room 127, just off the front lobby. Sessions available to attend in-person will be indicated in the announcement information. In-person participants without NASA badges will need a valid visitor badge (US citizenship or permanent residency is required) to enter the facility.

Join Us Online

To attend an AMS seminar, click on the Microsoft Teams link in the seminar description. You can connect through the desktop Teams App if you have it, or you can select "Watch on the Web Instead" to enter the meeting as a guest. Please ensure that you've muted your microphone in Teams before you select "Join Now" to enter the meeting lobby.