{"id":"https://openalex.org/W4388581204","doi":"https://doi.org/10.1145/3624062.3624153","title":"MPI-xCCL: A Portable MPI Library over Collective Communication Libraries for Various Accelerators","display_name":"MPI-xCCL: A Portable MPI Library over Collective Communication Libraries for Various Accelerators","publication_year":2023,"publication_date":"2023-11-10","ids":{"openalex":"https://openalex.org/W4388581204","doi":"https://doi.org/10.1145/3624062.3624153"},"language":"en","primary_location":{"id":"doi:10.1145/3624062.3624153","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3624062.3624153","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3624062.3624153","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the SC '23 Workshops of the International Conference on High Performance Computing, Network, Storage, and Analysis","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3624062.3624153","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5013189498","display_name":"Chen-Chun Chen","orcid":"https://orcid.org/0000-0002-7471-7552"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Chen-Chun Chen","raw_affiliation_strings":["The Ohio State University, United States of America"],"raw_orcid":"https://orcid.org/0000-0002-7471-7552","affiliations":[{"raw_affiliation_string":"The Ohio State University, United States of America","institution_ids":["https://openalex.org/I52357470"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5017245523","display_name":"Kawthar Shafie Khorassani","orcid":"https://orcid.org/0000-0001-5856-5483"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kawthar Shafie Khorassani","raw_affiliation_strings":["The Ohio State University, United States of America"],"raw_orcid":"https://orcid.org/0000-0001-5856-5483","affiliations":[{"raw_affiliation_string":"The Ohio State University, United States of America","institution_ids":["https://openalex.org/I52357470"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5057608531","display_name":"Pouya Kousha","orcid":"https://orcid.org/0009-0004-7507-0940"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Pouya Kousha","raw_affiliation_strings":["The Ohio State University, United States of America"],"raw_orcid":"https://orcid.org/0009-0004-7507-0940","affiliations":[{"raw_affiliation_string":"The Ohio State University, United States of America","institution_ids":["https://openalex.org/I52357470"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028296912","display_name":"Qinghua Zhou","orcid":"https://orcid.org/0009-0001-8490-7990"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Qinghua Zhou","raw_affiliation_strings":["The Ohio State University, United States of America"],"raw_orcid":"https://orcid.org/0009-0001-8490-7990","affiliations":[{"raw_affiliation_string":"The Ohio State University, United States of America","institution_ids":["https://openalex.org/I52357470"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015189372","display_name":"Jinghan Yao","orcid":"https://orcid.org/0009-0002-7129-9508"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jinghan Yao","raw_affiliation_strings":["The Ohio State University, United States of America"],"raw_orcid":"https://orcid.org/0009-0002-7129-9508","affiliations":[{"raw_affiliation_string":"The Ohio State University, United States of America","institution_ids":["https://openalex.org/I52357470"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5034293705","display_name":"Hari Subramoni","orcid":"https://orcid.org/0000-0002-1200-2754"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Hari Subramoni","raw_affiliation_strings":["The Ohio State University, USA"],"raw_orcid":"https://orcid.org/0000-0002-1200-2754","affiliations":[{"raw_affiliation_string":"The Ohio State University, USA","institution_ids":["https://openalex.org/I52357470"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5024879682","display_name":"Dhabaleswar K. Panda","orcid":"https://orcid.org/0000-0002-0356-1781"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Dhabaleswar K. Panda","raw_affiliation_strings":["The Ohio State University, USA"],"raw_orcid":"https://orcid.org/0000-0002-0356-1781","affiliations":[{"raw_affiliation_string":"The Ohio State University, USA","institution_ids":["https://openalex.org/I52357470"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I52357470"],"apc_list":null,"apc_paid":null,"fwci":2.537,"has_fulltext":true,"cited_by_count":7,"citation_normalized_percentile":{"value":0.90920343,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":91,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"847","last_page":"854"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11181","display_name":"Advanced Data Storage Technologies","score":0.9998000264167786,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11181","display_name":"Advanced Data Storage Technologies","score":0.9998000264167786,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9995999932289124,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11478","display_name":"Caching and Content Delivery","score":0.9976999759674072,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8783034682273865},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.745201587677002},{"id":"https://openalex.org/keywords/vendor","display_name":"Vendor","score":0.6301649808883667},{"id":"https://openalex.org/keywords/supercomputer","display_name":"Supercomputer","score":0.558786928653717},{"id":"https://openalex.org/keywords/cuda","display_name":"CUDA","score":0.5275737047195435},{"id":"https://openalex.org/keywords/message-passing-interface","display_name":"Message Passing Interface","score":0.5086844563484192},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.5067537426948547},{"id":"https://openalex.org/keywords/infiniband","display_name":"InfiniBand","score":0.503863513469696},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.5032784342765808},{"id":"https://openalex.org/keywords/usability","display_name":"Usability","score":0.46946656703948975},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.4647630453109741},{"id":"https://openalex.org/keywords/message-passing","display_name":"Message passing","score":0.4254189729690552},{"id":"https://openalex.org/keywords/gpu-cluster","display_name":"GPU cluster","score":0.4130750298500061},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.3983600437641144},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.09566399455070496}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8783034682273865},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.745201587677002},{"id":"https://openalex.org/C2777338717","wikidata":"https://www.wikidata.org/wiki/Q1762621","display_name":"Vendor","level":2,"score":0.6301649808883667},{"id":"https://openalex.org/C83283714","wikidata":"https://www.wikidata.org/wiki/Q121117","display_name":"Supercomputer","level":2,"score":0.558786928653717},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.5275737047195435},{"id":"https://openalex.org/C166782233","wikidata":"https://www.wikidata.org/wiki/Q127879","display_name":"Message Passing Interface","level":3,"score":0.5086844563484192},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.5067537426948547},{"id":"https://openalex.org/C2781030343","wikidata":"https://www.wikidata.org/wiki/Q922437","display_name":"InfiniBand","level":2,"score":0.503863513469696},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.5032784342765808},{"id":"https://openalex.org/C170130773","wikidata":"https://www.wikidata.org/wiki/Q216378","display_name":"Usability","level":2,"score":0.46946656703948975},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.4647630453109741},{"id":"https://openalex.org/C854659","wikidata":"https://www.wikidata.org/wiki/Q1859284","display_name":"Message passing","level":2,"score":0.4254189729690552},{"id":"https://openalex.org/C2781335571","wikidata":"https://www.wikidata.org/wiki/Q2633544","display_name":"GPU cluster","level":3,"score":0.4130750298500061},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.3983600437641144},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.09566399455070496},{"id":"https://openalex.org/C162853370","wikidata":"https://www.wikidata.org/wiki/Q39809","display_name":"Marketing","level":1,"score":0.0},{"id":"https://openalex.org/C144133560","wikidata":"https://www.wikidata.org/wiki/Q4830453","display_name":"Business","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3624062.3624153","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3624062.3624153","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3624062.3624153","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the SC '23 Workshops of the International Conference on High Performance Computing, Network, Storage, and Analysis","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3624062.3624153","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3624062.3624153","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3624062.3624153","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the SC '23 Workshops of the International Conference on High Performance Computing, Network, Storage, and Analysis","raw_type":"proceedings-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/9","display_name":"Industry, innovation and infrastructure","score":0.44999998807907104}],"awards":[{"id":"https://openalex.org/G202197749","display_name":"Computation for the Endless Frontier","funder_award_id":"1818253","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G3358531649","display_name":"Collaborative Research: Frameworks: Performance Engineering Scientific Applications with MVAPICH and TAU using Emerging Communication Primitives","funder_award_id":"2311830","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G3746232255","display_name":null,"funder_award_id":"NCR-130002","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G4196433103","display_name":"OAC Core: Small: Next-Generation Communication and I/O Middleware for HPC and Deep Learning with Smart NICs","funder_award_id":"2007991","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G5271487896","display_name":"Operations & Maintenance for the Endless Frontier","funder_award_id":"1854828","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G5279863110","display_name":"Collaborative Research: Frameworks: Designing Next-Generation MPI Libraries for Emerging Dense GPU Systems","funder_award_id":"1931537","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G8438779844","display_name":"CSR: Small: CONCERT: Designing Scalable Communication Runtimes with On-the-fly Compression for HPC and AI Applications on Heterogeneous Architectures","funder_award_id":"2312927","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G8585293727","display_name":"MRI: RADiCAL: Reconfigurable Major Research Cyberinfrastructure for Advanced Computational Data Analytics and Machine Learning","funder_award_id":"2018627","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4388581204.pdf","grobid_xml":"https://content.openalex.org/works/W4388581204.grobid-xml"},"referenced_works_count":8,"referenced_works":["https://openalex.org/W1637731592","https://openalex.org/W2025388037","https://openalex.org/W3166649773","https://openalex.org/W3177445289","https://openalex.org/W4284964190","https://openalex.org/W4289827868","https://openalex.org/W4362566357","https://openalex.org/W4383749601"],"related_works":["https://openalex.org/W2124048060","https://openalex.org/W4386915331","https://openalex.org/W2246013950","https://openalex.org/W2378910916","https://openalex.org/W2119634093","https://openalex.org/W2116006827","https://openalex.org/W1970720081","https://openalex.org/W2751263050","https://openalex.org/W2030707850","https://openalex.org/W2808583214"],"abstract_inverted_index":{"The":[0,87],"evolution":[1],"of":[2,23,94],"high-performance":[3],"computing":[4],"toward":[5],"diverse":[6],"accelerators,":[7],"including":[8],"NVIDIA,":[9],"AMD,":[10],"Intel":[11],"GPUs,":[12],"and":[13,20,31,67,96,99,116],"Habana":[14,135],"Gaudi":[15],"Accelerators,":[16],"demands":[17],"a":[18,55],"user-friendly":[19],"efficient":[21],"utilization":[22],"these":[24],"technologies.":[25],"While":[26],"both":[27],"GPU-aware":[28,76],"MPI":[29,77,85,95,130],"libraries":[30,34,74],"vendor-specific":[32,73],"communication":[33,37,69,148],"cater":[35],"to":[36,84,101,142],"requirements,":[38],"trade-offs":[39],"emerge":[40],"based":[41],"on":[42,109],"library":[43],"selection":[44],"across":[45],"various":[46,110],"message":[47],"sizes.":[48],"Thus,":[49],"prioritizing":[50],"usability,":[51],"we":[52],"propose":[53],"MPI-xCCL,":[54],"Message":[56],"Passing":[57],"Interface-based":[58],"runtime":[59],"with":[60,75,121],"cross-accelerator":[61],"support":[62],"for":[63],"efficient,":[64],"portable,":[65],"scalable,":[66],"optimized":[68],"performance.":[70],"MPI-xCCL":[71,138],"incorporates":[72],"runtimes":[78],"ensuring":[79],"multi-accelerator":[80],"compatibility":[81],"while":[82],"adhering":[83],"standards.":[86],"proposed":[88],"hybrid":[89],"designs":[90,108,127],"leverage":[91],"the":[92,102],"benefits":[93],"xCCL":[97],"algorithms":[98],"transparently":[100],"end":[103],"user.":[104],"We":[105],"evaluated":[106],"our":[107,126],"HPC":[111],"systems":[112],"using":[113],"OSU":[114],"Micro-Benchmarks,":[115],"Deep":[117],"Learning":[118],"frameworks":[119],"TensorFlow":[120],"Horovod.":[122],"On":[123,133],"NVIDIA-GPU-enabled":[124],"ThetaGPU,":[125],"outperformed":[128],"Open":[129],"by":[131],"4.6x.":[132],"emerging":[134],"Gaudi-based":[136],"systems,":[137],"was":[139],"also":[140],"able":[141],"deliver":[143],"similar":[144],"performance":[145],"as":[146],"vendor-provided":[147],"runtimes.":[149]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":3}],"updated_date":"2026-07-31T08:31:51.225901","created_date":"2025-10-10T00:00:00"}