| | |
Summary: Scalable Speculative Parallelization on Commodity Clusters
Hanjun Kim Arun Raman Feng Liu Jae W. Lee
David I. August
Departments of Electrical Engineering and Computer Science Parakinetics Inc.
Princeton University Princeton, USA
Princeton, USA leejw@parakinetics.com
{hanjunk, rarun, fengliu, august}@princeton.edu
Abstract
While clusters of commodity servers and switches are the most
popular form of large-scale parallel computers, many programs are
not easily parallelized for execution upon them. In particular, high
inter-node communication cost and lack of globally shared memory
appear to make clusters suitable only for server applications with
abundant task-level parallelism and scientific applications with regu-
lar and independent units of work. Clever use of pipeline parallelism
(DSWP), thread-level speculation (TLS), and speculative pipeline
parallelism (Spec-DSWP) can mitigate the costs of inter-thread
communication on shared memory multicore machines. This paper
presents Distributed Software Multi-threaded Transactional memory
(DSMTX), a runtime system which makes these techniques applicable
|
