Minimal Multithreading - Exploiting Redundancy in Parallel Systems

Speaker:        Dr. Diana FRANKLIN
                University of California at Santa Barbara

Title:          "Minimal Multithreading - Exploiting Redundancy in
                Parallel Systems"

Date:           Wednesday, 7 December 2011

Time:           4:00pm - 5:00pm

Venue:          Lecture Theatre H (near lifts 27/28), HKUST

Abstract:

Parallelism and energy efficiency have emerged as critical considerations
in future computing systems.  We observe that many convenient models for
expressing parallelism can lead to significant redundancy in computation
and data. Our research explores a set of microarchitectural mechanisms
that support these software models while effectively eliminating this
redundancy, providing high performance and energy efficiency.
Specifically, we observe that many parallel software models (e.g. data
parallelism, Single-Program Multiple Data (SPMD), multi-programming, and
high-throughput computing) consist of multiple threads of computation with
very similar instructions streams and working sets of data. The trick,
however, is to design efficient mechanisms that both exploit this
similarity and effectively support the differences.

We describe two research thrusts to eliminate redundancy across several
application domains. First, content-aware caching techniques that reduce
redundant storage of identical data across parallel threads and processes.
Second, multi-threaded processor core designs that eliminate redundant
instruction fetch and/or execution.

***********************
Biography:

Diana Franklin is tenured teaching faculty and Director of the Center for
Computing Education and Diversity at UCSB.  Franklin received her Ph.D.
from UC Davis in 2002 and is a recipient of the NSF CAREER award. She was
an assistant professor (2002-2007) and associate professor (2007) of
Computer Science at the California Polytechnic State University, during
which she held the Forbes Chair (2002-2007).  Her research interests
include parallel programming and architecture, computing education, and
ethnic and gender diversity in computing.