{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,22]],"date-time":"2026-01-22T22:31:53Z","timestamp":1769121113577,"version":"3.49.0"},"reference-count":61,"publisher":"IEEE","license":[{"start":{"date-parts":[[2022,4,1]],"date-time":"2022-04-01T00:00:00Z","timestamp":1648771200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2022,4,1]],"date-time":"2022-04-01T00:00:00Z","timestamp":1648771200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/100006602","name":"Air Force Research Laboratory","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100006602","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2022,4]]},"DOI":"10.1109\/hpca53966.2022.00093","type":"proceedings-article","created":{"date-parts":[[2022,5,17]],"date-time":"2022-05-17T19:50:17Z","timestamp":1652817017000},"page":"1224-1237","source":"Crossref","is-referenced-by-count":37,"title":["AI-Enabling Workloads on Large-Scale GPU-Accelerated System: Characterization, Opportunities, and Implications"],"prefix":"10.1109","author":[{"given":"Baolin","family":"Li","sequence":"first","affiliation":[{"name":"Northeastern University"}]},{"given":"Rohin","family":"Arora","sequence":"additional","affiliation":[{"name":"Northeastern University"}]},{"given":"Siddharth","family":"Samsi","sequence":"additional","affiliation":[{"name":"MIT Lincoln Laboratory"}]},{"given":"Tirthak","family":"Patel","sequence":"additional","affiliation":[{"name":"Northeastern University"}]},{"given":"William","family":"Arcand","sequence":"additional","affiliation":[{"name":"MIT Lincoln Laboratory"}]},{"given":"David","family":"Bestor","sequence":"additional","affiliation":[{"name":"MIT Lincoln Laboratory"}]},{"given":"Chansup","family":"Byun","sequence":"additional","affiliation":[{"name":"MIT Lincoln Laboratory"}]},{"given":"Rohan Basu","family":"Roy","sequence":"additional","affiliation":[{"name":"Northeastern University"}]},{"given":"Bill","family":"Bergeron","sequence":"additional","affiliation":[{"name":"MIT Lincoln Laboratory"}]},{"given":"John","family":"Holodnak","sequence":"additional","affiliation":[{"name":"MIT Lincoln Laboratory"}]},{"given":"Michael","family":"Houle","sequence":"additional","affiliation":[{"name":"MIT Lincoln Laboratory"}]},{"given":"Matthew","family":"Hubbell","sequence":"additional","affiliation":[{"name":"MIT Lincoln Laboratory"}]},{"given":"Michael","family":"Jones","sequence":"additional","affiliation":[{"name":"MIT Lincoln Laboratory"}]},{"given":"Jeremy","family":"Kepner","sequence":"additional","affiliation":[{"name":"MIT Lincoln Laboratory"}]},{"given":"Anna","family":"Klein","sequence":"additional","affiliation":[{"name":"MIT Lincoln Laboratory"}]},{"given":"Peter","family":"Michaleas","sequence":"additional","affiliation":[{"name":"MIT Lincoln Laboratory"}]},{"given":"Joseph","family":"McDonald","sequence":"additional","affiliation":[{"name":"MIT Lincoln Laboratory"}]},{"given":"Lauren","family":"Milechin","sequence":"additional","affiliation":[{"name":"MIT Lincoln Laboratory"}]},{"given":"Julie","family":"Mullen","sequence":"additional","affiliation":[{"name":"MIT Lincoln Laboratory"}]},{"given":"Andrew","family":"Prout","sequence":"additional","affiliation":[{"name":"MIT Lincoln Laboratory"}]},{"given":"Benjamin","family":"Price","sequence":"additional","affiliation":[{"name":"MIT Lincoln Laboratory"}]},{"given":"Albert","family":"Reuther","sequence":"additional","affiliation":[{"name":"MIT Lincoln Laboratory"}]},{"given":"Antonio","family":"Rosa","sequence":"additional","affiliation":[{"name":"MIT Lincoln Laboratory"}]},{"given":"Matthew","family":"Weiss","sequence":"additional","affiliation":[{"name":"MIT Lincoln Laboratory"}]},{"given":"Charles","family":"Yee","sequence":"additional","affiliation":[{"name":"MIT Lincoln Laboratory"}]},{"given":"Daniel","family":"Edelman","sequence":"additional","affiliation":[{"name":"MIT Lincoln Laboratory"}]},{"given":"Allan","family":"Vanterpool","sequence":"additional","affiliation":[{"name":"US Air Force"}]},{"given":"Anson","family":"Cheng","sequence":"additional","affiliation":[{"name":"US Air Force"}]},{"given":"Vijay","family":"Gadepally","sequence":"additional","affiliation":[{"name":"MIT Lincoln Laboratory"}]},{"given":"Devesh","family":"Tiwari","sequence":"additional","affiliation":[{"name":"Northeastern University"}]}],"member":"263","reference":[{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/WORKS49585.2019.00009"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/IPDPS47924.2020.00087"},{"key":"ref33","first-page":"8024","article-title":"Pytorch: An imperative style, high-performance deep learning library","author":"paszke","year":"2019","journal-title":"Advances in Neural IInformation Processing Systems"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/SC41405.2020.00045"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1145\/1254882.1254939"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/DSN.2018.00022"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA47549.2020.00025"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1145\/3307681.3326607"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/SC41405.2020.00088"},{"key":"ref34","first-page":"103","article-title":"{GIFT}: A coupon based throttleand-reward mechanism for fair and efficient i\/o bandwidth management on parallel storage systems","author":"patel","year":"2020","journal-title":"18th USENIX Conference on File and Storage Technologies ( FAST 20)"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1007\/10968987_3"},{"key":"ref61","first-page":"515","article-title":"HiveD: Sharing a {GPU} Cluster for Deep Learning with Guarantees","author":"zhao","year":"2020","journal-title":"OSDI 2020 14th USENIX Symposium on Operating Systems Design and Implementation"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.3390\/jlpea8020013"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/GLOBECOM38437.2019.9014110"},{"key":"ref29","first-page":"481","article-title":"Heterogeneity-Aware Cluster Scheduling Policies for Deep Learning Workloads","author":"narayanan","year":"2020","journal-title":"OSDI 2020 14th USENIX Symposium on Operating Systems Design and Implementation"},{"key":"ref2","first-page":"469","article-title":"Cherrypick: Adaptively Unearthing the Best Cloud Configurations for Big Data Analytics","author":"alipourfard","year":"2017","journal-title":"14th USENIX Symposium on Networked Systems Design and Implementation NSDI 17)"},{"key":"ref1","first-page":"265","article-title":"Tensorflow: A system for largescale machine learning","author":"abadi","year":"2016","journal-title":"12th USENIX Symposium on Operating Systems Design and Implementation ( OSDI 16)"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/CCGrid49817.2020.00-66"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1145\/3458817.3476223"},{"key":"ref21","article-title":"Deep residual learning for image recognition","author":"he","year":"2015"},{"key":"ref24","article-title":"Workload Analysis of Blue Waters","author":"jones","year":"2017"},{"key":"ref23","first-page":"947","article-title":"Analysis of Large-Scale Multi-Tenant {GPU} Clusters for {DNN} Training Workloads","author":"jeon","year":"2019","journal-title":"2019 USENIX Annual Technical Conference ( USENIX ATC 19)"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/DSN-S.2019.00020"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1145\/2304576.2304611"},{"key":"ref50","article-title":"Very deep convolutional networks for large-scale image recognition","author":"simonyan","year":"2015"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1177\/1094342014522573"},{"key":"ref59","article-title":"Practical Resilience Analysis of GPGPU Applications in the Presence of Single-and Multi-bit Faults","author":"yang","year":"2020","journal-title":"IEEE Transactions on Computers"},{"key":"ref58","first-page":"595","article-title":"Gandiva: Introspective Cluster Scheduling for Deep Learning","author":"xiao","year":"2018","journal-title":"13th USENIX Symposium on Operating Systems Design and Implementation ( OSDI 18)"},{"key":"ref57","first-page":"9","article-title":"What’s Working in HPC: Investigating HPC User Behavior and Productivity","volume":"2","author":"wolter","year":"2006","journal-title":"CTWatch Quarterly"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.22369\/issn.2153-4136\/10\/1\/9"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1145\/2807591.2807666"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA.2015.7056044"},{"key":"ref53","first-page":"111","article-title":"Challenges to the Supercomputer Development in Russia: a HPC User Perspective","volume":"5","author":"stegailov","year":"2014","journal-title":"Program Systems Theory and Applications"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1145\/3311790.3396656"},{"key":"ref10","article-title":"Nvidia-smi","author":"corporation","year":"2016"},{"key":"ref11","article-title":"Dcgm","author":"corporation","year":"2021"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/DSN.2011.5958210"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1145\/3400286.3418263"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1145\/3419111.3421284"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/DSN.2014.62"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1145\/3203217.3203273"},{"key":"ref16","article-title":"Pytorch lightning","volume":"3","author":"falcon","year":"2019","journal-title":"Github"},{"key":"ref17","author":"feldman","year":"2018","journal-title":"Top500"},{"key":"ref18","first-page":"485","article-title":"Tiresias: A {GPU} Cluster Manager for Distributed Deep Learning","author":"gu","year":"2019","journal-title":"16th USENIX Symposium on Networked Systems Design and Implementation ( NSDI 19)"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1145\/3126908.3126937"},{"key":"ref4","first-page":"521","article-title":"Towards Understanding Job Heterogeneity in HPC: A NERSC Case Study","author":"alvarez","year":"2016","journal-title":"2016 16th IEEE\/ACM International Symposium on Cluster Cloud and Grid Computing (CCGrid)"},{"key":"ref3","first-page":"1","article-title":"Experience and Practice of Batch Scheduling on Leadership Supercomputers at Argonne","author":"allcock","year":"2017","journal-title":"Workshop on Job Scheduling Strategies for Parallel Processing"},{"key":"ref6","article-title":"Language models are few-shot learners","author":"brown","year":"2020"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1016\/j.cpc.2010.11.011"},{"key":"ref8","doi-asserted-by":"crossref","first-page":"107","DOI":"10.1145\/3297858.3304005","article-title":"Parties: Qos-Aware Resource Partitioning for Multiple Interactive Services","author":"chen","year":"2019","journal-title":"Proceedings of the fourth international conference on Architectural support for programming languages and operating systems - AS"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1145\/3342195.3387555"},{"key":"ref49","article-title":"A Workload Analysis of NSF’s Innovative HPC Resources Using XDMoD","author":"simakov","year":"2018"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/MM.2021.3061394"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1145\/2907294.2907314"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/HPCSim.2016.7568362"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/ICDCS.2017.259"},{"key":"ref47","article-title":"Horovod: fast and easy distributed deep learning in TensorFlow","author":"sergeev","year":"2018"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/HPEC43674.2020.9286149"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/HPEC.2018.8547629"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1109\/HPEC49654.2021.9622850"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1016\/j.jpdc.2017.09.002"}],"event":{"name":"2022 IEEE International Symposium on High-Performance Computer Architecture (HPCA)","location":"Seoul, Korea, Republic of","start":{"date-parts":[[2022,4,2]]},"end":{"date-parts":[[2022,4,6]]}},"container-title":["2022 IEEE International Symposium on High-Performance Computer Architecture (HPCA)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/9773179\/9773180\/09773216.pdf?arnumber=9773216","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,6,20]],"date-time":"2022-06-20T21:34:52Z","timestamp":1655760892000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9773216\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,4]]},"references-count":61,"URL":"https:\/\/doi.org\/10.1109\/hpca53966.2022.00093","relation":{},"subject":[],"published":{"date-parts":[[2022,4]]}}}