{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,30]],"date-time":"2026-05-30T12:01:35Z","timestamp":1780142495289,"version":"3.54.0"},"reference-count":64,"publisher":"Elsevier BV","license":[{"start":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:00:00Z","timestamp":1785542400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/tdm\/userlicense\/1.0\/"},{"start":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:00:00Z","timestamp":1785542400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/legal\/tdmrep-license"},{"start":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:00:00Z","timestamp":1785542400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-017"},{"start":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:00:00Z","timestamp":1785542400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"},{"start":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:00:00Z","timestamp":1785542400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-012"},{"start":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:00:00Z","timestamp":1785542400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:00:00Z","timestamp":1785542400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-004"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62272476"],"award-info":[{"award-number":["62272476"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100002855","name":"Ministry of Science and Technology of the People's Republic of China","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100002855","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100012164","name":"National High-tech Research and Development Program","doi-asserted-by":"publisher","award":["2023YFB3001504"],"award-info":[{"award-number":["2023YFB3001504"]}],"id":[{"id":"10.13039\/501100012164","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["elsevier.com","sciencedirect.com"],"crossmark-restriction":true},"short-container-title":["Journal of Parallel and Distributed Computing"],"published-print":{"date-parts":[[2026,8]]},"DOI":"10.1016\/j.jpdc.2026.105274","type":"journal-article","created":{"date-parts":[[2026,4,14]],"date-time":"2026-04-14T13:16:03Z","timestamp":1776172563000},"page":"105274","update-policy":"https:\/\/doi.org\/10.1016\/elsevier_cm_policy","source":"Crossref","is-referenced-by-count":0,"special_numbering":"C","title":["Self -adaptive and topology-aware broadcast leveraging collective offload on Tianhe express interconnect"],"prefix":"10.1016","volume":"214","author":[{"ORCID":"https:\/\/orcid.org\/0009-0005-0489-9548","authenticated-orcid":false,"given":"Chongshan","family":"Liang","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yi","family":"Dai","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xingyun","family":"Qi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jiaqing","family":"Xu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Min","family":"Xie","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dongsheng","family":"Li","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Weixia","family":"Xu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"78","reference":[{"key":"10.1016\/j.jpdc.2026.105274_bib0001","series-title":"IEEE International Conference on High Performance Computing, Data, and Analytics","article-title":"Characterization of MPI usage on a production supercomputer","author":"Chunduri","year":"2018"},{"key":"10.1016\/j.jpdc.2026.105274_bib0002","series-title":"2009 IEEE International Symposium on Parallel & Distributed Processing IPDPS","article-title":"Designing multi-leader-based allgather algorithms for multi-core clusters","author":"Kandalla","year":"2009"},{"key":"10.1016\/j.jpdc.2026.105274_bib0003","series-title":"2018 IEEE International Conference on Cluster Computing (CLUSTER)","first-page":"12","article-title":"Salar: scalable and adaptive designs for large message reduction collectives","author":"Bayatpour","year":"2018"},{"key":"10.1016\/j.jpdc.2026.105274_bib0004","series-title":"2008 IEEE International Symposium on Parallel and Distributed Processing","first-page":"1","article-title":"Scaling alltoall collective on multi-core systems","author":"Kumar","year":"2008"},{"key":"10.1016\/j.jpdc.2026.105274_bib0005","series-title":"SC23: International Conference for High Performance Computing, Networking, Storage and Analysis","article-title":"Optimizing MPI collectives on shared memory multi-Cores","author":"Peng","year":"2023"},{"key":"10.1016\/j.jpdc.2026.105274_bib0006","series-title":"25th IEEE International Conference on Cluster Computing (CLUSTER)","article-title":"Uniform algorithms for reduce-scatter and (most) other collectives for MPI","author":"Traeff","year":"2023"},{"key":"10.1016\/j.jpdc.2026.105274_bib0007","series-title":"2015 IEEE 23rd Annual Symposium on High-Performance Interconnects (HOTI)","article-title":"Intel omni-path architecture: enabling scalable, high performance fabrics","author":"Birrittella","year":"2015"},{"key":"10.1016\/j.jpdc.2026.105274_bib0008","series-title":"22nd USENIX Symposium on Networked Systems Design and Implementation (NSDI 25)","first-page":"705","article-title":"Efficient direct-connect topologies for collective communications","author":"Zhao","year":"2025"},{"key":"10.1016\/j.jpdc.2026.105274_bib0009","series-title":"2016 IEEE International Parallel and Distributed Processing Symposium (IPDPS)","article-title":"Optimization and analysis of MPI collective communication on fat-tree networks","author":"Kumar","year":"2016"},{"key":"10.1016\/j.jpdc.2026.105274_bib0010","series-title":"36th ACM International Conference on Supercomputing, ICS 2022","article-title":"Optimized MPI collective algorithms for dragonfly topology","author":"Feng","year":"2022"},{"key":"10.1016\/j.jpdc.2026.105274_bib0011","series-title":"High Performance Computing","article-title":"BluesMPI: efficient MPI non-blocking alltoall offloading designs on modern bluefield smart NICs","author":"Bayatpour","year":"2021"},{"key":"10.1016\/j.jpdc.2026.105274_bib0012","series-title":"IEEE International Conference on Cluster Computing","first-page":"477","article-title":"Designing an offloaded nonblocking MPI_allgather collective using CORE-direct","author":"Inozemtsev","year":"2012"},{"key":"10.1016\/j.jpdc.2026.105274_bib0013","series-title":"SC \u201911: Proceedings of 2011 International Conference for High Performance Computing, Networking, Storage and Analysis","first-page":"1","article-title":"The IBM blue Gene\/Q interconnection network and message unit","author":"Chen","year":"2011"},{"issue":"3","key":"10.1016\/j.jpdc.2026.105274_bib0014","doi-asserted-by":"crossref","first-page":"575","DOI":"10.1109\/TPDS.2018.2867222","article-title":"Exploiting hardware multicast and GPUDirect RDMA for efficient broadcast","volume":"30","author":"Chu","year":"2019","journal-title":"IEEE Trans. Parallel Distrib. Syst."},{"key":"10.1016\/j.jpdc.2026.105274_bib0015","series-title":"2016 First International Workshop on Communication Optimizations in HPC (COMHPC)","article-title":"Scalable hierarchical aggregation protocol (sharp): a hardware architecture for efficient data reduction","author":"Graham","year":"2016"},{"key":"10.1016\/j.jpdc.2026.105274_bib0016","doi-asserted-by":"crossref","first-page":"70","DOI":"10.1016\/j.future.2023.10.010","article-title":"Canary: congestion-aware in-network allreduce using dynamic trees","volume":"152","author":"De Sensi","year":"2024","journal-title":"Future Gener. Comput. Syst."},{"key":"10.1016\/j.jpdc.2026.105274_bib0017","series-title":"Third IEEE International Symposium on Network Computing and Applications, 2004. (NCA 2004). Proceedings.","first-page":"323","article-title":"Myrinet networks: A performance study","author":"Qian","year":"2004"},{"key":"10.1016\/j.jpdc.2026.105274_bib0018","series-title":"2011 IEEE 19th Annual Symposium on High Performance Interconnects","first-page":"35","article-title":"Enabling flexible collective communication offload with triggered operations","author":"Underwood","year":"2011"},{"key":"10.1016\/j.jpdc.2026.105274_bib0019","series-title":"The Proceedings of Hot Chips\u2019 03","article-title":"QsNet-II: an interconnect for supercomputing applications","author":"Beecroft","year":"2003"},{"key":"10.1016\/j.jpdc.2026.105274_bib0020","series-title":"Proceedings 15th International Parallel and Distributed Processing Symposium. IPDPS 2001","first-page":"8","article-title":"Fast NIC-based barrier over Myrinet\/GM","author":"Buntinas","year":"2001"},{"key":"10.1016\/j.jpdc.2026.105274_bib0021","series-title":"18th International Parallel and Distributed Processing Symposium, 2004. Proceedings.","first-page":"182","article-title":"Efficient and scalable barrier over quadrics and myrinet with a new nic-based collective message passing protocol","author":"Yu","year":"2004"},{"key":"10.1016\/j.jpdc.2026.105274_bib0022","series-title":"International Workshop on Communication, Architecture, and Applications for Network-Based Parallel Computing","first-page":"115","article-title":"Broadcast\/multicast over Myrinet using NIC-assisted multidestination messages","author":"Buntinas","year":"2000"},{"issue":"1","key":"10.1016\/j.jpdc.2026.105274_bib0023","doi-asserted-by":"crossref","first-page":"49","DOI":"10.1177\/1094342005051521","article-title":"Optimization of collective communication operations in MPICH","volume":"19","author":"Thakur","year":"2005","journal-title":"Int. J. High Perform. Comput. Appl."},{"key":"10.1016\/j.jpdc.2026.105274_bib0024","series-title":"IPDRM 2019","article-title":"Leveraging network-level parallelism with multiple process-endpoints for MPI broadcast","author":"Ruhela","year":"2019"},{"key":"10.1016\/j.jpdc.2026.105274_bib0025","series-title":"High Performance Interconnects","article-title":"Designing non-blocking broadcast with collective offload on InfiniBand clusters: a case study with HPL","author":"Kandalla","year":"2011"},{"key":"10.1016\/j.jpdc.2026.105274_bib0026","series-title":"2021 IEEE 28th International Conference on High Performance Computing, Data, and Analytics (HiPC)","first-page":"388","article-title":"Large-message nonblocking MPI_iallgather and MPI ibcast offload via BlueField-2\u202fDPU","author":"Sarkauskas","year":"2021"},{"key":"10.1016\/j.jpdc.2026.105274_bib0027","series-title":"18th International Parallel and Distributed Processing Symposium, 2004. Proceedings.","article-title":"Fast and scalable MPI-level broadcast using InfiniBand\u2019s hardware multicast support","author":"Liu","year":"2004"},{"issue":"23","key":"10.1016\/j.jpdc.2026.105274_bib0028","first-page":"41","article-title":"InfiniBand trade association announces RDMA over converged ethernet (RoCE)","year":"2010","journal-title":"ANSI Stand. Actions"},{"key":"10.1016\/j.jpdc.2026.105274_bib0029","series-title":"International Conference on Parallel & Distributed Processing","article-title":"Efficient SMP-aware MPI-level broadcast over InfiniBand\u2019s hardware multicast","author":"Mamidala","year":"2006"},{"key":"10.1016\/j.jpdc.2026.105274_bib0030","series-title":"Cray User Group 2012 Conference","article-title":"Leveraging the cray linux environment core specialization feature to realize MPI asynchronous progress on cray XE systems","author":"Pritchard","year":"2012"},{"key":"10.1016\/j.jpdc.2026.105274_bib0031","series-title":"2008 IEEE International Symposium on Parallel and Distributed Processing","first-page":"1","article-title":"A multithreaded communication engine for multicore architectures","author":"Trahay","year":"2008"},{"issue":"3-4","key":"10.1016\/j.jpdc.2026.105274_bib0032","doi-asserted-by":"crossref","first-page":"237","DOI":"10.1007\/s00450-011-0170-4","article-title":"High-performance and scalable non-blocking all-to-all with collective offload on InfiniBand clusters: a study with parallel 3D FFT","volume":"26","author":"Kandalla","year":"2011","journal-title":"Comput. Science - Res. Develop."},{"key":"10.1016\/j.jpdc.2026.105274_bib0033","doi-asserted-by":"crossref","first-page":"13","DOI":"10.1016\/j.parco.2019.03.003","article-title":"Efficient design for MPI asynchronous progress without dedicated resources","volume":"85","author":"Ruhela","year":"2019","journal-title":"Parallel Comput."},{"key":"10.1016\/j.jpdc.2026.105274_bib0034","series-title":"2024 IEEE International Parallel and Distributed Processing Symposium (IPDPS)","article-title":"The self-adaptive and topology-aware MPI_bcast leveraging collective offload on Tianhe express interconnect","author":"Liang","year":"2024"},{"key":"10.1016\/j.jpdc.2026.105274_bib0035","doi-asserted-by":"crossref","first-page":"259","DOI":"10.1007\/s11390-015-1520-7","article-title":"High performance interconnect network for Tianhe system","volume":"30","author":"Liao","year":"2015","journal-title":"J. Comput. Sci. Technol."},{"key":"10.1016\/j.jpdc.2026.105274_bib0036","doi-asserted-by":"crossref","first-page":"237","DOI":"10.1007\/s00450-011-0170-4","article-title":"High-performance and scalable non-blocking all-to-all with collective offload on InfiniBand clusters: a study with parallel 3D FFT","volume":"26","author":"Kandalla","year":"2011","journal-title":"Comput. Science - Research and Dev."},{"issue":"9","key":"10.1016\/j.jpdc.2026.105274_bib0037","doi-asserted-by":"crossref","first-page":"624","DOI":"10.1016\/j.parco.2007.06.006","article-title":"Optimizing a conjugate gradient solver with non-blocking collective operations","volume":"33","author":"Hoefler","year":"2007","journal-title":"Parallel Comput."},{"key":"10.1016\/j.jpdc.2026.105274_bib0038","unstructured":"Argonne National Laboratory, The progress engine - MPICH, 2018, (In wiki). https:\/\/www.mpich.org\/mpich\/index.php\/ The Progress Engine."},{"key":"10.1016\/j.jpdc.2026.105274_bib0039","series-title":"Workshop on Sustainable Software for Science: Practice and Experiences, Held in Conjunction with Int\u2019l Conference on Supercomputing (WSSPE)","article-title":"The MVAPICH project: evolution and sustainability of an open source production quality MPI library for HPC","author":"Panda","year":"2013"},{"key":"10.1016\/j.jpdc.2026.105274_bib0040","series-title":"2006 IEEE International Conference on Cluster Computing","first-page":"1","article-title":"Open MPI: a high-performance, heterogeneous MPI","author":"Graham","year":"2006"},{"key":"10.1016\/j.jpdc.2026.105274_bib0041","series-title":"Proceedings of the 2006 International Conference on Frontiers of High Performance Computing and Networking","first-page":"155","article-title":"A case for non-blocking collective operations","author":"Hoefler","year":"2006"},{"key":"10.1016\/j.jpdc.2026.105274_bib0042","series-title":"2011 IEEE 19th Annual Symposium on High Performance Interconnects","article-title":"Designing non-blocking broadcast with collective offload on InfiniBand clusters: a case study with HPL","author":"Kandalla","year":"2011"},{"key":"10.1016\/j.jpdc.2026.105274_bib0043","series-title":"Technology Brief","article-title":"CORE-Direct: The Most Advanced Technology for MPI\/SHMEM Collectives Offloads","year":"2010"},{"key":"10.1016\/j.jpdc.2026.105274_bib0044","series-title":"2021 IEEE 28th International Conference on High Performance Computing, Data, and Analytics (HiPC)","first-page":"388","article-title":"Large-message nonblocking mpi_iallgather and mpi_ibcast offload via bluefield-2 dpu","author":"Sarkauskas","year":"2021"},{"key":"10.1016\/j.jpdc.2026.105274_bib0045","unstructured":"OSU micro-benchmarks, 2026, (http:\/\/mvapich.cse.ohio-state.edu\/benchmarks\/)."},{"key":"10.1016\/j.jpdc.2026.105274_bib0046","doi-asserted-by":"crossref","unstructured":"R. Zambre, M. Grodowitz, A. Chandramowlishwaran, P. Shamis, Breaking Band: A Breakdown of High-performance Communication, arXiv e-prints (2020).","DOI":"10.1145\/3337821.3337910"},{"issue":"2","key":"10.1016\/j.jpdc.2026.105274_bib0047","doi-asserted-by":"crossref","first-page":"259","DOI":"10.1007\/s11390-015-1520-7","article-title":"High performance interconnect network for Tianhe system","author":"Liao","year":"2015","journal-title":"J. Comput. Sci. Technol."},{"issue":"7","key":"10.1016\/j.jpdc.2026.105274_bib0048","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1007\/s11432-015-5292-z","article-title":"The sunway TaihuLight supercomputer: system and applications","volume":"59","author":"Jin","year":"2016","journal-title":"Sci. China Inf. Sci."},{"key":"10.1016\/j.jpdc.2026.105274_bib0049","unstructured":"Tianhe-2A, 2021, (https:\/\/www.top500.org\/system\/177999\/). Accessed: 2025-07-28."},{"key":"10.1016\/j.jpdc.2026.105274_bib0050","unstructured":"MPICH, 2022, https:\/\/www.mpich.org\/."},{"key":"10.1016\/j.jpdc.2026.105274_bib0051","unstructured":"OpenMPI, 2022, https:\/\/www.open-mpi.org\/."},{"key":"10.1016\/j.jpdc.2026.105274_bib0052","unstructured":"SPEC MPI, 2021, https:\/\/www.spec.org\/mpi2007."},{"key":"10.1016\/j.jpdc.2026.105274_bib0053","series-title":"International Conference for High Performance Computing","article-title":"Scalable reduction collectives with data partitioning-based multi-leader design","author":"Bayatpour","year":"2017"},{"key":"10.1016\/j.jpdc.2026.105274_bib0054","series-title":"2022 IEEE International Parallel and Distributed Processing Symposium (IPDPS)","first-page":"1305","article-title":"The fast and scalable MPI application launch of the Tianhe HPC system","author":"Dai","year":"2022"},{"key":"10.1016\/j.jpdc.2026.105274_bib0055","series-title":"IEEE International Conference on Cluster Computing","article-title":"Design and implementation of broadcast algorithms for extreme-scale systems","author":"Shamis","year":"2011"},{"key":"10.1016\/j.jpdc.2026.105274_bib0056","series-title":"2008 IEEE International Symposium on Parallel and Distributed Processing IPDPS","article-title":"Efficient MPI bcast across different process arrival patterns","author":"Patarasuk","year":"2008"},{"key":"10.1016\/j.jpdc.2026.105274_bib0057","series-title":"2015 44th International Conference on Parallel Processing Workshops","first-page":"111","article-title":"A bandwidth-saving optimization for mpi broadcast collective operation","author":"Zhou","year":"2015"},{"key":"10.1016\/j.jpdc.2026.105274_bib0058","doi-asserted-by":"crossref","first-page":"70","DOI":"10.1016\/j.parco.2018.05.006","article-title":"Hardware topology management in MPI applications through hierarchical communicators","volume":"76","author":"Goglin","year":"2018","journal-title":"Parallel Comput."},{"issue":"11","key":"10.1016\/j.jpdc.2026.105274_bib0059","doi-asserted-by":"crossref","first-page":"3079","DOI":"10.1109\/TPDS.2013.96","article-title":"Network performance aware MPI collective communication operations in the cloud","volume":"26","author":"Gong","year":"2015","journal-title":"IEEE Trans. Parallel Distrib. Syst."},{"key":"10.1016\/j.jpdc.2026.105274_bib0060","series-title":"International Conference on High Performance Computing","article-title":"Designing non-blocking personalized collectives with near perfect overlap for RDMA-enabled clusters","author":"Subramoni","year":"2015"},{"key":"10.1016\/j.jpdc.2026.105274_bib0061","series-title":"International Conference on High Performance Computing","first-page":"334","article-title":"Designing dynamic and adaptive MPI point-to-point communication protocols for efficient overlap of computation and communication","author":"Subramoni","year":"2017"},{"key":"10.1016\/j.jpdc.2026.105274_bib0062","series-title":"2008 IEEE International Conference on Cluster Computing","first-page":"213","article-title":"Message progression in parallel computing-to thread or not to thread?","author":"Hoefler","year":"2008"},{"key":"10.1016\/j.jpdc.2026.105274_bib0063","series-title":"Proceedings of the Cray User Group Conference","first-page":"130","article-title":"Leveraging the cray linux environment core specialization feature to realize mpi asynchronous progress on cray xe systems","volume":"79","author":"Pritchard","year":"2012"},{"key":"10.1016\/j.jpdc.2026.105274_bib0064","series-title":"Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis","first-page":"1","article-title":"Improving concurrency and asynchrony in multithreaded MPI applications using software offloading","author":"Vaidyanathan","year":"2015"}],"container-title":["Journal of Parallel and Distributed Computing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0743731526000523?httpAccept=text\/xml","content-type":"text\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0743731526000523?httpAccept=text\/plain","content-type":"text\/plain","content-version":"vor","intended-application":"text-mining"}],"deposited":{"date-parts":[[2026,5,30]],"date-time":"2026-05-30T11:10:52Z","timestamp":1780139452000},"score":1,"resource":{"primary":{"URL":"https:\/\/linkinghub.elsevier.com\/retrieve\/pii\/S0743731526000523"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8]]},"references-count":64,"alternative-id":["S0743731526000523"],"URL":"https:\/\/doi.org\/10.1016\/j.jpdc.2026.105274","relation":{},"ISSN":["0743-7315"],"issn-type":[{"value":"0743-7315","type":"print"}],"subject":[],"published":{"date-parts":[[2026,8]]},"assertion":[{"value":"Elsevier","name":"publisher","label":"This article is maintained by"},{"value":"Self -adaptive and topology-aware broadcast leveraging collective offload on Tianhe express interconnect","name":"articletitle","label":"Article Title"},{"value":"Journal of Parallel and Distributed Computing","name":"journaltitle","label":"Journal Title"},{"value":"https:\/\/doi.org\/10.1016\/j.jpdc.2026.105274","name":"articlelink","label":"CrossRef DOI link to publisher maintained version"},{"value":"article","name":"content_type","label":"Content Type"},{"value":"\u00a9 2026 Elsevier Inc. All rights are reserved, including those for text and data mining, AI training, and similar technologies.","name":"copyright","label":"Copyright"}],"article-number":"105274"}}