{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T02:09:55Z","timestamp":1785463795227,"version":"3.56.0"},"reference-count":40,"publisher":"SAGE Publications","issue":"4","license":[{"start":{"date-parts":[[2021,5,13]],"date-time":"2021-05-13T00:00:00Z","timestamp":1620864000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/journals.sagepub.com\/page\/policies\/text-and-data-mining-license"}],"content-domain":{"domain":["journals.sagepub.com"],"crossmark-restriction":true},"short-container-title":["The International Journal of High Performance Computing Applications"],"published-print":{"date-parts":[[2021,7]]},"abstract":"<jats:p>With increasing complexity of HPC workflows, data management services need to perform expensive I\/O operations asynchronously in the background, aiming to overlap the I\/O with the application runtime. However, this may cause interference due to competition for resources: CPU, memory\/network bandwidth. The advent of multi-core architectures has exacerbated this problem, as many I\/O operations are issued concurrently, thereby competing not only with the application but also among themselves. Furthermore, the interference patterns can dynamically change as a response to variations in application behavior and I\/O subsystems (e.g. multiple users sharing a parallel file system). Without a thorough understanding, I\/O operations may perform suboptimally, potentially even worse than in the blocking case. To fill this gap, this paper investigates the causes and consequences of interference due to asynchronous I\/O on HPC systems. Specifically, we focus on multi-core CPUs and memory bandwidth, isolating the interference due to each resource. Then, we perform an in-depth study to explain the interplay and contention in a variety of resource sharing scenarios such as varying priority and number of background I\/O threads and different I\/O strategies: sendfile, read\/write, mmap\/write underlining trade-offs. The insights from this study are important both to enable guided optimizations of existing background I\/O, as well as to open new opportunities to design advanced asynchronous I\/O strategies.<\/jats:p>","DOI":"10.1177\/10943420211016511","type":"journal-article","created":{"date-parts":[[2021,5,13]],"date-time":"2021-05-13T05:46:04Z","timestamp":1620884764000},"page":"391-412","update-policy":"https:\/\/doi.org\/10.1177\/sage-journals-update-policy","source":"Crossref","is-referenced-by-count":19,"title":["Demystifying asynchronous I\/O Interference in HPC applications"],"prefix":"10.1177","volume":"35","author":[{"given":"Shu-Mei","family":"Tseng","sequence":"first","affiliation":[{"name":"EECS, University of California Irvine, California, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0661-7509","authenticated-orcid":false,"given":"Bogdan","family":"Nicolae","sequence":"additional","affiliation":[{"name":"MCS, Argonne National Laboratory, Illinois, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Franck","family":"Cappello","sequence":"additional","affiliation":[{"name":"MCS, Argonne National Laboratory, Illinois, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Aparna","family":"Chandramowlishwaran","sequence":"additional","affiliation":[{"name":"EECS, University of California Irvine, California, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"179","published-online":{"date-parts":[[2021,5,13]]},"reference":[{"key":"bibr1-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1002\/cpe.1631"},{"key":"bibr2-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1007\/s41781-019-0026-3"},{"key":"bibr3-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1109\/MSST.2012.6232376"},{"key":"bibr4-10943420211016511","doi-asserted-by":"crossref","unstructured":"Bercea GT, McRae AT, Ham DA, et al. (2016) A structure-exploiting numbering algorithm for finite elements on extruded meshes, and its performance evaluation in Firedrake.\n                      arXiv preprint arXiv:1604.05937\n                      .","DOI":"10.5194\/gmd-9-3803-2016"},{"key":"bibr5-10943420211016511","first-page":"1","volume-title":"HPC \u201817: The 25th High Performance Computing Symposium","author":"Cao L","year":"2017"},{"key":"bibr6-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1145\/2027066.2027068"},{"key":"bibr7-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1109\/IPPS.1999.760432"},{"key":"bibr8-10943420211016511","first-page":"89","volume":"13","author":"Dongarra JJ","year":"1997","journal-title":"Supercomputer"},{"key":"bibr9-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1142\/S0129626411000151"},{"key":"bibr10-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1109\/SC.2014.77"},{"key":"bibr11-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1109\/IPDPS.2015.116"},{"key":"bibr12-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1109\/IPDPS.2016.29"},{"key":"bibr13-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1145\/2503210.2504566"},{"key":"bibr14-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1109\/IPDPSW.2018.00127"},{"key":"bibr15-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2018.2867879"},{"key":"bibr16-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1145\/165854.165874"},{"key":"bibr17-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1016\/j.future.2018.05.051"},{"key":"bibr18-10943420211016511","volume-title":"Spacehog: evaluating the costs of dedicating resources to in situ analysis","author":"Kreitinger R","year":"2017"},{"key":"bibr19-10943420211016511","doi-asserted-by":"publisher","DOI":"10.2172\/1632124"},{"key":"bibr20-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1109\/SC.2016.49"},{"key":"bibr21-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1109\/SC.2010.32"},{"key":"bibr22-10943420211016511","first-page":"19","volume-title":"IEEE Computer Society Technical Committee on Computer Architecture (TCCA) Newsletter","author":"McCalpin JD","year":"1995"},{"key":"bibr23-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1109\/IPDPS.2019.00099"},{"key":"bibr24-10943420211016511","volume-title":"Early evaluation of the Cray XC40 Xeon Phi system \u2018Theta\u2019 at Argonne","author":"Parker S","year":"2017"},{"key":"bibr25-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1109\/BigData.2015.7363979"},{"key":"bibr26-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1109\/IPDPSW.2016.24"},{"key":"bibr27-10943420211016511","volume-title":"Inside the Lustre file system. Technical report","author":"Petersen TK","year":"2015"},{"key":"bibr28-10943420211016511","first-page":"1","volume-title":"HPEC\u201913: The 2013 IEEE High Performance Extreme Computing Conference","author":"Petersen TK","year":"2013"},{"key":"bibr29-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1109\/HPEC.2014.7040948"},{"key":"bibr30-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1145\/2699414"},{"key":"bibr31-10943420211016511","volume-title":"Storage systems and input\/output to support extreme scale science. Technical report","author":"Ross R","year":"2015"},{"key":"bibr32-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1007\/s11227-016-1904-7"},{"key":"bibr33-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1109\/HiPC.2019.00023"},{"key":"bibr34-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1109\/CCGRID.2018.00026"},{"key":"bibr35-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-58667-0_17"},{"key":"bibr36-10943420211016511","first-page":"1","volume-title":"SC\u201910: The 2010 International Conference for High Performance Computing, Networking, Storage and Analysis","author":"Vishwanath V","year":"2010"},{"key":"bibr37-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1109\/IPDPS.2016.50"},{"key":"bibr38-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1145\/1810085.1810116"},{"key":"bibr39-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1145\/2503210.2503279"},{"key":"bibr40-10943420211016511","doi-asserted-by":"publisher","DOI":"10.1109\/CLUSTER.2015.45"}],"container-title":["The International Journal of High Performance Computing Applications"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/10943420211016511","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/full-xml\/10.1177\/10943420211016511","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/10943420211016511","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T08:17:12Z","timestamp":1777450632000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/10.1177\/10943420211016511"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,5,13]]},"references-count":40,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2021,7]]}},"alternative-id":["10.1177\/10943420211016511"],"URL":"https:\/\/doi.org\/10.1177\/10943420211016511","relation":{},"ISSN":["1094-3420","1741-2846"],"issn-type":[{"value":"1094-3420","type":"print"},{"value":"1741-2846","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,5,13]]}}}