דלג לתוכן (מקש קיצור 's')
אירועים

אירועים והרצאות בפקולטה למדעי המחשב ע"ש הנרי ומרילין טאוב

event speaker icon
אשיש ונקט (אונ' סאן דייגו)
event date icon
יום רביעי, 09.11.2016, 11:30
event location icon
טאוב 401
On-chip heterogeneity has been shown to be an effective mechanism to improve execution efficiency for general purpose and embedded computing. Existing heterogeneous designs either feature a single ISA or multiple ISAs that statically partition work. In this talk, PhD candidate from UC San Diego Ashish Venkat, will describe his research that enables programs to cross a heretofore forbidden boundary -- the ISA. In particular, Venkat will describe a compiler and runtime strategy for swift and seamless execution migration across diverse ISAs. Early design space exploration from Venkat's research shows that harnessing ISA diversity provides substantial performance gains and energy savings by allowing a program to freely migrate across heterogeneous-ISA cores during different phases of its execution, and equips chip architects with finer design choices that enable more efficient ISA-microarchitecture co-design. In addition, Venkat will also briefly discuss his recent work that demonstrates the immense security potential of cross-ISA migration to thwart buffer overflow based attacks such as Return-Oriented Programming. Finally, he will briefly touch upon his ongoing collaboration at HRL under the EU OPERA project. which plans to leverage the aforementioned techniques to explore the potential efficiency benefits of ISA-heterogeneity in a datacenter environment.

Bio:
Ashish Venkat is a PhD Candidate in the Computer Science and Engineering department at UC San Diego, where he is a member of the High Performance Processor Architecture and Compilation lab, directed by Prof. Dean Tullsen. His research interests are in computer architecture and compilers, especially in instruction set design, processor microarchitecture, binary translation, code generation, and their intersection with computer security and machine learning. His work on Heterogeneous-ISA Chip Multiprocessors has been published at ISCA and ASPLOS.