Sergio Rivas-Gomez, A. Fanfarillo, Sai B. Narasimhamurthy, S. Markidis
{"title":"持久队列:在队列fortran中集成MPI存储窗口","authors":"Sergio Rivas-Gomez, A. Fanfarillo, Sai B. Narasimhamurthy, S. Markidis","doi":"10.1145/3343211.3343214","DOIUrl":null,"url":null,"abstract":"The inherent integration of novel hardware and software components on HPC is expected to considerably aggravate the Mean Time Between Failures (MTBF) on scientific applications, while simultaneously increase the programming complexity of these clusters. In this work, we present the initial steps towards the integration of transparent resilience support inside Coarray Fortran. In particular, we propose persistent coarrays, an extension of OpenCoarrays that integrates MPI storage windows to leverage its transport layer and seamlessly map coarrays to files on storage. Preliminary results indicate that our approach provides clear benefits on representative workloads, while incurring in minimal source code changes.","PeriodicalId":314904,"journal":{"name":"Proceedings of the 26th European MPI Users' Group Meeting","volume":"1943 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2019-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"4","resultStr":"{\"title\":\"Persistent coarrays: integrating MPI storage windows in coarray fortran\",\"authors\":\"Sergio Rivas-Gomez, A. Fanfarillo, Sai B. Narasimhamurthy, S. Markidis\",\"doi\":\"10.1145/3343211.3343214\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The inherent integration of novel hardware and software components on HPC is expected to considerably aggravate the Mean Time Between Failures (MTBF) on scientific applications, while simultaneously increase the programming complexity of these clusters. In this work, we present the initial steps towards the integration of transparent resilience support inside Coarray Fortran. In particular, we propose persistent coarrays, an extension of OpenCoarrays that integrates MPI storage windows to leverage its transport layer and seamlessly map coarrays to files on storage. Preliminary results indicate that our approach provides clear benefits on representative workloads, while incurring in minimal source code changes.\",\"PeriodicalId\":314904,\"journal\":{\"name\":\"Proceedings of the 26th European MPI Users' Group Meeting\",\"volume\":\"1943 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2019-09-11\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"4\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Proceedings of the 26th European MPI Users' Group Meeting\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1145/3343211.3343214\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the 26th European MPI Users' Group Meeting","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/3343211.3343214","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Persistent coarrays: integrating MPI storage windows in coarray fortran
The inherent integration of novel hardware and software components on HPC is expected to considerably aggravate the Mean Time Between Failures (MTBF) on scientific applications, while simultaneously increase the programming complexity of these clusters. In this work, we present the initial steps towards the integration of transparent resilience support inside Coarray Fortran. In particular, we propose persistent coarrays, an extension of OpenCoarrays that integrates MPI storage windows to leverage its transport layer and seamlessly map coarrays to files on storage. Preliminary results indicate that our approach provides clear benefits on representative workloads, while incurring in minimal source code changes.