Name
Abstract | ||
|---|---|---|
Support Vector Machines (SVMs) are used to discover method-specific compilation strategies in Testarossa, a commercial Just-in-Time (JiT) compiler employed in the IBM® J9 Java™ Virtual Machine. The learning process explores a large number of different compilation strategies to generate the data needed for training models. The trained machine-learned model is integrated with the compiler to predict a compilation plan that balances code quality and compilation effort on a per-method basis. The machine-learned plans outperform the original Testarossa for start-up performance, but not for throughput performance, for which Testarossa has been highly hand-tuned for many years.
|
Year | DOI | Venue |
|---|---|---|
2011 | 10.1109/CGO.2011.5764693 | CGO |
Keywords | Field | DocType |
trained machine-learned model,method-specific compilation strategy,machine-learned plan,different compilation strategy,j9 java,original testarossa,throughput performance,start-up performance,compilation effort,compilation plan,radiation detectors,optimization,virtual machine,support vector machines,java,learning artificial intelligence,just in time compiler,support vector machine,code quality,machine learning,data models,virtual machines | Data modeling,Virtual machine,Programming language,Dynamic compilation,Computer science,Support vector machine,Compiler,Real-time computing,Just-in-time compilation,Tracing just-in-time compilation,Software quality | Conference |
ISSN | ISBN | Citations |
2164-2397 | 978-1-61284-356-8 | 5 |
PageRank | References | Authors |
0.45 | 13 | 5 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Ricardo Nabinger Sanchez | 1 | 7 | 0.84 |
| Jose Nelson Amaral | 2 | 70 | 12.40 |
| D. Szafron | 3 | 1579 | 210.88 |
| Marius Pirvu | 4 | 9 | 2.23 |
| Mark Stoodley | 5 | 119 | 8.71 |