From:                     Kendra Smith

Sent:                      Saturday, May 13, 2000 12:54 AM

To:                         M?crosöft Research Tech Talk, Sem. Notice

Cc:                         Kendra Smith

Subject:                 UW-CSE Colloq / 5-24-2000 / Ruf / M?crosöft / Sy nchronization Elimination via Thread Closure and Summary-Based Flow Analyses

UW-CSE Colloq / 5-24-2000 / Ruf / M?crosöft / Sy nchronization Elimination via Thread Closure and Summary-Based Flow Analyses

 

*NOTE*  This lecture will not be videotaped.

 

UNIVERSITY OF WASHINGTON

Seattle, Washington 98195

 

Department of Computer Science and Engineering

Box 352350

(206) 543-1695

 

COLLOQUIUM

 

SPEAKER:      Erik Ruf, M?crosöft

 

TITLE:          Synchronization Elimination via Thread Closure and

                             Summary-Based Flow Analyses

 

DATE:           Wednesday, May 24, 2000

 

TIME:           2:30pm

 

PLACE:                   EE1-003

 

HOST:           Susan Eggers

 

ABSTRACT:

 

We present a new technique for removing unnecessary synchronization

operations from statically compiled Java programs.  Our approach

improves upon current efforts based on escape analysis, as it can

eliminate synchronization operations even on objects that escape

their allocating threads. It makes use of a compact,

equivalence-class-based representation that eliminates the need for

fixed point operations during the analysis.

 

We describe and evaluate the performance of an implementation in the

Marmot native Java compiler.  For the benchmark programs examined, the

optimization removes 100\% of the dynamic synchronization operations

in single-threaded programs, and 0-99\% in multi-threaded programs, at

a low cost in additional compilation time and code growth.  We also

discuss benefits and limitations of the general approach, in both the

specific context of synchronization elimination and in the more

general context of value flow based transformations.

 

BIO

 

Erik Ruf joined M?crosöft Research in 1993 and has been a member of

the Advanced Programming Languages group since 1997.  His interests

include sparse program representations, efficient program analyses,

and automated specialization of both programs and data.  His current

work focuses on performance optimizations for the Marmot Java

compiler.  Erik holds a B.S. degree in Computer Engineering from Case

Western Reserve University and M.S. and Ph.D. degrees in Electrical

Engineering from Stanford University.

 

 

Email: talk-info@cs.washington.edu

Info: http://www.cs.washington.edu