Performance search engine driven by prior knowledge of optimization
Kim, Y., Černý, P., Dennis, J. M.. (2015). Performance search engine driven by prior knowledge of optimization. ARRAY 2015: Proceedings of the 2nd ACM SIGPLAN International Workshop on Libraries, Languages, and Compilers for Array Programming, doi:10.1145/2774959.2774963
| Title | Performance search engine driven by prior knowledge of optimization |
|---|---|
| Author(s) | Youngsung Kim, Pavol Černý, John M. Dennis |
| Abstract | For scientific array-based programs, optimization for a particular target platform is a hard problem. There are many optimization techniques such as (semantics-preserving) source code transformations, compiler directives, environment variables, and compiler flags that influence performance. Moreover, the performance impact of (combinations of) these factors is unpredictable. This pa- per focuses on providing a platform for automatically searching through search space consisting of such optimization techniques. We provide (i) a search-space description language, which enables the user to describe optimization options to be used; (ii) search engine that enables testing the performance impact of optimization options by executing optimized programs and checking their results; and (iii) an interface for implementing various search algorithms. We evaluate our platform by using two simple search algorithms - a random search and a casetree search that heuristically learns from the already examined parts of the search space. We show that such algorithms are easily implementable in our plat- form, and we empirically find that the framework can be used to find useful optimized algorithms. |
| Publication Title | ARRAY 2015: Proceedings of the 2nd ACM SIGPLAN International Workshop on Libraries, Languages, and Compilers for Array Programming |
| Publication Date | Jul 1, 2015 |
| Publisher's Version of Record | https://dx.doi.org/10.1145/2774959.2774963 |
| OpenSky Citable URL | https://n2t.net/ark:/85065/d7g44s2s |
| OpenSky Listing | View on OpenSky |
| CISL Affiliations | TDD, ASAP |