DocumentCode :
2582836
Title :
Harmony: Collection and analysis of parallel block vectors
Author :
Kambadur, Melanie ; Tang, Kui ; Kim, Martha A.
Author_Institution :
Columbia Univ., New York, NY, USA
fYear :
2012
fDate :
9-13 June 2012
Firstpage :
452
Lastpage :
463
Abstract :
Efficient execution of well-parallelized applications is central to performance in the multicore era. Program analysis tools support the hardware and software sides of this effort by exposing relevant features of multithreaded applications. This paper describes parallel block vectors, which uncover previously unseen characteristics of parallel programs. Parallel block vectors provide block execution profiles per concurrency phase (e.g., the block execution profile of all serial regions of a program). This information provides a direct and fine-grained mapping between an application´s runtime parallel phases and the static code that makes up those phases. This paper also demonstrates how to collect parallel block vectors with minimal application perturbation using Harmony. Harmony is an instrumentation pass for the LLVM compiler that introduces just 16-21% overhead on average across eight Parsec benchmarks. We apply parallel block vectors to uncover several novel insights about parallel applications with direct consequences for architectural design. First, that the serial and parallel phases of execution used in Amdahl´s Law are often composed of many of the same basic blocks. Second, that program features, such as instruction mix, vary based on the degree of parallelism, with serial phases in particular displaying different instruction mixes from the program as a whole. Third, that dynamic execution frequencies do not necessarily correlate with a block´s parallelism.
Keywords :
concurrency control; multi-threading; program compilers; program diagnostics; software architecture; software tools; Harmony; LLVM compiler; Parsec benchmarks; architectural design; block execution profiles; concurrency phase; direct mapping; dynamic execution frequencies; fine-grained mapping; hardware support; instruction mix; minimal application perturbation; multithreaded applications; parallel block vector analysis; parallel block vector collection; parallel block vectors; parallel programs; program analysis tools; runtime parallel phases; serial execution phases; software support; static code; Instruction sets; Instruments; Parallel processing; Radiation detectors; Registers; Runtime; Vectors;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computer Architecture (ISCA), 2012 39th Annual International Symposium on
Conference_Location :
Portland, OR
ISSN :
1063-6897
Print_ISBN :
978-1-4673-0475-7
Electronic_ISBN :
1063-6897
Type :
conf
DOI :
10.1109/ISCA.2012.6237039
Filename :
6237039
Link To Document :
بازگشت