October 15th, 2014

Anton: A Specialized Supercomputer for Molecular Dynamic Solutions

Marty Deneroff

Chief Operating Office
Emu Solutions

 

Anton was conceived as a mechanism to perform molecular simulations at a scale and fidelity not achievable by any other means. The talk will outline the reasons these simulations are so difficult, the motivations for performing them, and the approaches used to build a device that is able to successfully meet these challenges.

About Marty Deneroff

Marty Deneroff recently joined Emu Solutions as Chief Operating Officer. Marty is a veteran technology executive with a wealth of experience in high performance computing. Past positions include Head of Engineering at D.E. Shaw Research in New York and Vice President, Server and Microprocessor Engineering at SGI in Mountain View California. While at D.E. Shaw, Marty led the development of Anton, a special purpose computer that performs molecular dynamic simulations at unprecedented rates. While at SGI, which is widely considered an industry leader in high performance computing, Marty led the development and subsequent support for all of SGI's server and microprocessor products, including Origin an Altix servers and MIPS R14000, R16000 and R18000 microprocessors. Marty was also the project head and chief architect for SGI’s entry into DARPA HPCS program, known as Ultraviolet. This was a petaflop-scale multiprocessor system capable of practical and highly productive use across a broad range of applications.

In addition to his industry leadership experience, Marty is also the author of a several publications and currently holds 18 patents in the field of computer architecture and design. In 2009, The Association for Computer Machinery awarded Marty the Gordon Bell Prize for Special Achievement. The prize recognizes outstanding achievement in high-performance computing applications. The purpose of the award is to track the progress over time of parallel computing, with particular emphasis on rewarding innovation in applying high-performance computing to applications in science, engineering, and large-scale data analytics. Prior to beginning his professional career, Marty earned a Bachelor's of Science in Electrical Engineering from MIT and a Master's of Science in Computer Science from Monmouth University.