%0 Conference Paper %1 Hayashi:2015:LCO %A Hayashi, Akihiro %A Zhao, Jisheng %A Ferguson, Michael %A Sarkar, Vivek %B Proceedings of the Second Workshop on the LLVM Compiler Infrastructure in HPC %D 2015 %I ACM %K LLVM Locality Merging MessageSending Optimizations PGAS %P 1:1--1:11 %R 10.1145/2833157.2833164 %T LLVM-based Communication Optimizations for PGAS Programs %X While Partitioned Global Address Space (PGAS) programming languages such as UPC/UPC++, CAF, Chapel and X10 provide high-level programming models for facilitating large-scale distributed-memory parallel programming, it is widely recognized that compiler analysis and optimization for these languages has been very limited, unlike the optimization of SMP models such as OpenMP. One reason for this limitation is that current optimizers for PGAS programs are specialized to different languages. This is unfortunate since communication optimization is an important class of compiler optimizations for PGAS programs running on distributed-memory platforms, and these optimizations need to be performed more widely. Thus, a more effective approach would be to build a language-independent and runtime-independent compiler framework for optimizing PGAS programs so that new communication optimizations can be leveraged by different languages. To address this need, we introduce an LLVM-based (Low Level Virtual Machine) communication optimization framework. Our compilation system leverages existing optimization passes and introduces new PGAS language-aware runtime dependent/independent passes to reduce communication overheads. Our experimental results show an average performance improvement of 3.5× and 3.4× on 64-nodes of a Cray XC30TM supercomputer and 32-nodes of a Westmere cluster respectively, for a set of benchmarks written in the Chapel language. Overall, we show that our new LLVM-based compiler optimization framework can effectively improve the performance of PGAS programs. %@ 978-1-4503-4005-2

@inproceedings{Hayashi:2015:LCO, abstract = {While Partitioned Global Address Space (PGAS) programming languages such as UPC/UPC++, CAF, Chapel and X10 provide high-level programming models for facilitating large-scale distributed-memory parallel programming, it is widely recognized that compiler analysis and optimization for these languages has been very limited, unlike the optimization of SMP models such as OpenMP. One reason for this limitation is that current optimizers for PGAS programs are specialized to different languages. This is unfortunate since communication optimization is an important class of compiler optimizations for PGAS programs running on distributed-memory platforms, and these optimizations need to be performed more widely. Thus, a more effective approach would be to build a language-independent and runtime-independent compiler framework for optimizing PGAS programs so that new communication optimizations can be leveraged by different languages. To address this need, we introduce an LLVM-based (Low Level Virtual Machine) communication optimization framework. Our compilation system leverages existing optimization passes and introduces new PGAS language-aware runtime dependent/independent passes to reduce communication overheads. Our experimental results show an average performance improvement of 3.5× and 3.4× on 64-nodes of a Cray XC30TM supercomputer and 32-nodes of a Westmere cluster respectively, for a set of benchmarks written in the Chapel language. Overall, we show that our new LLVM-based compiler optimization framework can effectively improve the performance of PGAS programs.}, acmid = {2833164}, added-at = {2015-11-25T12:10:54.000+0100}, articleno = {1}, author = {Hayashi, Akihiro and Zhao, Jisheng and Ferguson, Michael and Sarkar, Vivek}, biburl = {https://www.bibsonomy.org/bibtex/21d3dd29b3b3e092368b68954906f4076/gron}, booktitle = {Proceedings of the Second Workshop on the LLVM Compiler Infrastructure in HPC}, description = {LLVM-based communication optimizations for PGAS programs}, doi = {10.1145/2833157.2833164}, interhash = {9ba1f98cb9faa174886722b867ae00b8}, intrahash = {1d3dd29b3b3e092368b68954906f4076}, isbn = {978-1-4503-4005-2}, keywords = {LLVM Locality Merging MessageSending Optimizations PGAS}, location = {Austin, Texas}, numpages = {11}, pages = {1:1--1:11}, publisher = {ACM}, series = {LLVM '15}, timestamp = {2015-11-25T12:10:54.000+0100}, title = {LLVM-based Communication Optimizations for PGAS Programs}, year = 2015 }