Tue Feb 20, 2007, 11:00-12:00, 4750 Boelter Hall

X10: A High-Productivity Approach to Programming Multi-Core Systems

Vivek Sarkar
IBM Watson

It is now well established that current and future computer systems
are rapidly moving from single-core to multi-core processors so as to
use parallelism instead of frequency scaling as the foundation for
increased compute capacity. Unlike previous generations of hardware
evolution, this shift will have a major impact on existing
software. Current language facilities for concurrent programming are
inadequate for addressing the needs of multi-core systems because they
do not support general parallel structures combined with non-uniform
models of data access.

We have designed a modern object-oriented language, X10 (x10.sf.net),
for high performance, high productivity parallel programming. A member
of the partitioned global address space family of languages, X10
highlights the explicit reification of locality in the form of places;
lightweight activities embodied in async, future, foreach, and ateach;
constructs; a construct for termination detection (finish); the use of
lock-free synchronization (atomic blocks); and the manipulation of
inter-place global data structures. We present an overview of the X10
programming model and language, experiences with our reference
implementation, and discuss future directions for both homogeneous and
heterogeneous multi-core architectures.

This is joint work with other members of the X10 core team, including Vijay
Saraswat, Rajkishore Barik, Philippe Charles , Chris Donawa, Robert
Fuhrer, Allan Kielstra, Nate Nystrom, Igor Peshansky, Christoph von
Praun, Tong Wen.

BIO:

Vivek Sarkar is Senior Manager of the Programming Technologies
Department in the IBM Research Division, a member of the IBM Academy
of Technology, and an ACM Distinguished Scientist. The projects under
way in his department span the areas of 1) Programming Models and
Programming Language Design: Collage, DALI/XJ, X10, 2) Development
Tools: Continuous Software Quality (includes SAFE, Security Analysis,
Scripting Analysis), PERCS Parallel Tools, SAFARI, 3) Deployment,
Execution, Optimization: Dynamic Optimization, Jikes RVM, Metronome,,
PDS/Mirage.  Vivek's responsibilities also include leading the
research efforts in Programming Model, Tools, and Productivity in the
IBM PERCS project as part of the DARPA-funded program on High
Productivity Computing Systems.

Vivek's personal research interest is in advancing the state of the
art in high-level programming language design and implementation so as
to fully exploit the latest advances in processor hardware.  This
includes advances in programming tools, optimizing compilers, and
parallelizing compilers to deliver improved productivity and,
performance benefits to users of high-performance systems.  In the
past, Vivek has worked on several compiler-related projects at IBM
including the PTRAN and Jikes RVM research projects at Watson, and the
ASTI project at IBM Santa Teresa Laboratory which provided the
foundation for high-level locality optimizations and OpenMP and HPF
parallelization in IBM's XL Fortran product compilers.  Vivek holds a
B.Tech. degree from the Indian Institute of Technology, Kanpur, an
M.S. degree from University of Wisconsin-Madison, and a Ph.D. from
Stanford University. In 1997, he was on sabbatical as a visiting
associate professor at MIT, where he was a founding member of the MIT
RAW multi-core project.

Host: Jens Palsberg