Just-in-Time Compilation-Inspired Methodology for Parallelization of Compute Intensive Java Code
Compute intensive programs generally consume significant fraction of execution time in a small amount of repetitive code. Such repetitive code is commonly known as hotspot code. We observed that compute intensive hotspots often possess exploitable loop level parallelism. A JIT (Just-in-Time) compile...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Mehran University of Engineering and Technology
2017-01-01
|
| Series: | Mehran University Research Journal of Engineering and Technology |
| Subjects: | |
| Online Access: | http://publications.muet.edu.pk/research_papers/pdf/pdf1450.pdf |
| _version_ | 1819095110267174912 |
|---|---|
| author | GHULAM MUSTAFA WAQAR MAHMOOD MUHAMMAD USMAN GHANI |
| author_facet | GHULAM MUSTAFA WAQAR MAHMOOD MUHAMMAD USMAN GHANI |
| author_sort | GHULAM MUSTAFA |
| collection | DOAJ |
| description | Compute intensive programs generally consume significant fraction of execution time in a small amount of repetitive code. Such repetitive code is commonly known as hotspot code. We observed that compute intensive hotspots often possess exploitable loop level parallelism. A JIT (Just-in-Time) compiler profiles a running program to identify its hotspots. Hotspots are then translated into native code, for efficient execution. Using similar approach, we propose a methodology to identify hotspots and exploit their parallelization potential on multicore systems. Proposed methodology selects and parallelizes each DOALL loop that is either contained in a hotspot method or calls a hotspot method. The methodology could be integrated in front-end of a JIT compiler to parallelize sequential code, just before native translation. However, compilation to native code is out of scope of this work. As a case study, we analyze eighteen JGF (Java Grande Forum) benchmarks to determine parallelization potential of hotspots. Eight benchmarks demonstrate a speedup of up to 7.6x on an 8-core system |
| first_indexed | 2024-12-21T23:38:05Z |
| format | Article |
| id | doaj.art-2b540e168ad44efcbeec7a6bec73c467 |
| institution | Directory Open Access Journal |
| issn | 0254-7821 2413-7219 |
| language | English |
| last_indexed | 2024-12-21T23:38:05Z |
| publishDate | 2017-01-01 |
| publisher | Mehran University of Engineering and Technology |
| record_format | Article |
| series | Mehran University Research Journal of Engineering and Technology |
| spelling | doaj.art-2b540e168ad44efcbeec7a6bec73c4672022-12-21T18:46:19ZengMehran University of Engineering and TechnologyMehran University Research Journal of Engineering and Technology0254-78212413-72192017-01-0136167881450Just-in-Time Compilation-Inspired Methodology for Parallelization of Compute Intensive Java CodeGHULAM MUSTAFAWAQAR MAHMOODMUHAMMAD USMAN GHANICompute intensive programs generally consume significant fraction of execution time in a small amount of repetitive code. Such repetitive code is commonly known as hotspot code. We observed that compute intensive hotspots often possess exploitable loop level parallelism. A JIT (Just-in-Time) compiler profiles a running program to identify its hotspots. Hotspots are then translated into native code, for efficient execution. Using similar approach, we propose a methodology to identify hotspots and exploit their parallelization potential on multicore systems. Proposed methodology selects and parallelizes each DOALL loop that is either contained in a hotspot method or calls a hotspot method. The methodology could be integrated in front-end of a JIT compiler to parallelize sequential code, just before native translation. However, compilation to native code is out of scope of this work. As a case study, we analyze eighteen JGF (Java Grande Forum) benchmarks to determine parallelization potential of hotspots. Eight benchmarks demonstrate a speedup of up to 7.6x on an 8-core systemhttp://publications.muet.edu.pk/research_papers/pdf/pdf1450.pdfJust-in-Time CompilationLoop Level ParallelizationMulticore SystemRuntime AnalysisJava Virtual Machine. |
| spellingShingle | GHULAM MUSTAFA WAQAR MAHMOOD MUHAMMAD USMAN GHANI Just-in-Time Compilation-Inspired Methodology for Parallelization of Compute Intensive Java Code Mehran University Research Journal of Engineering and Technology Just-in-Time Compilation Loop Level Parallelization Multicore System Runtime Analysis Java Virtual Machine. |
| title | Just-in-Time Compilation-Inspired Methodology for Parallelization of Compute Intensive Java Code |
| title_full | Just-in-Time Compilation-Inspired Methodology for Parallelization of Compute Intensive Java Code |
| title_fullStr | Just-in-Time Compilation-Inspired Methodology for Parallelization of Compute Intensive Java Code |
| title_full_unstemmed | Just-in-Time Compilation-Inspired Methodology for Parallelization of Compute Intensive Java Code |
| title_short | Just-in-Time Compilation-Inspired Methodology for Parallelization of Compute Intensive Java Code |
| title_sort | just in time compilation inspired methodology for parallelization of compute intensive java code |
| topic | Just-in-Time Compilation Loop Level Parallelization Multicore System Runtime Analysis Java Virtual Machine. |
| url | http://publications.muet.edu.pk/research_papers/pdf/pdf1450.pdf |
| work_keys_str_mv | AT ghulammustafa justintimecompilationinspiredmethodologyforparallelizationofcomputeintensivejavacode AT waqarmahmood justintimecompilationinspiredmethodologyforparallelizationofcomputeintensivejavacode AT muhammadusmanghani justintimecompilationinspiredmethodologyforparallelizationofcomputeintensivejavacode |