| | |
Summary: 8
Performance Scalability of Decoupled
Software Pipelining
RAM RANGAN
IBM Austin Research Laboratory
and
NEIL VACHHARAJANI, GUILHERME OTTONI, and DAVID I. AUGUST
Princeton University
Any successful solution to using multicore processors to scale general-purpose program
performance will have to contend with rising intercore communication costs while exposing coarse-
grained parallelism. Recently proposed pipelined multithreading (PMT) techniques have been
demonstrated to have general-purpose applicability and are also able to effectively tolerate inter-
core latencies through pipelined interthread communication. These desirable properties make PMT
techniques strong candidates for program parallelization on current and future multicore proces-
sors and understanding their performance characteristics is critical to their deployment. To that
end, this paper evaluates the performance scalability of a general-purpose PMT technique called
decoupled software pipelining (DSWP) and presents a thorough analysis of the communication
bottlenecks that must be overcome for optimal DSWP scalability.
Categories and Subject Descriptors: C.4 [Performance of Systems]: Design Studies, Performance
Attributes; D.1.2 [Automatic Programming]: Program Transformation; D.1.3 [Concurrent Pro-
|
