{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,25]],"date-time":"2026-10-25T04:34:55Z","timestamp":1792902895333,"version":"4.5.0"},"reference-count":67,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"1","funder":[{"DOI":"10.13039\/501100000266","name":"Engineering and Physical Sciences Research Council","doi-asserted-by":"publisher","award":["EP\/P031617\/1"],"award-info":[{"award-number":["EP\/P031617\/1"]}],"id":[{"id":"10.13039\/501100000266","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Parallel Distrib. Syst."],"published-print":{"date-parts":[[2022,1,1]]},"DOI":"10.1109\/tpds.2021.3079202","type":"journal-article","created":{"date-parts":[[2021,5,11]],"date-time":"2021-05-11T15:49:42Z","timestamp":1620748182000},"page":"88-100","source":"Crossref","is-referenced-by-count":99,"title":["Horus: Interference-Aware and Prediction-Based Scheduling in Deep Learning Systems"],"prefix":"10.1109","volume":"33","author":[{"given":"Gingfung","family":"Yeung","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Damian","family":"Borowiec","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Renyu","family":"Yang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Adrian","family":"Friday","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Richard","family":"Harper","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Peter","family":"Garraghan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Highly scalable deep learning training system with mixed-precision: Training ImageNet in four minutes","author":"Jia","year":"2018"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/MM.2018.022071131"},{"key":"ref3","article-title":"Cloud GPUs | Google cloud","year":"2020"},{"key":"ref4","article-title":"Amazon EC2 P3 \u2013 Ideal for machine learning and HPC - AWS","year":"2020"},{"key":"ref5","article-title":"Azure VM sizes - GPU - Azure virtual machines","year":"2020"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1145\/3341301.3359658"},{"key":"ref7","first-page":"1049","article-title":"MArk: Exploiting cloud services for cost-effective, SLO-aware machine learning inference serving","volume-title":"Proc. USENIX Conf. Usenix Annu. Tech. Conf.","author":"Zhang"},{"key":"ref8","first-page":"595","article-title":"Gandiva: Introspective cluster scheduling for deep learning","volume-title":"Proc. 13th USENIX Conf. Operating Syst. Des. Implementation","author":"Xiao"},{"key":"ref9","first-page":"947","article-title":"Analysis of large-scale multi-tenant GPU clusters for DNN training workloads","volume-title":"Proc. USENIX Conf. Usenix Annu. Tech. Conf.","author":"Jeon"},{"key":"ref10","first-page":"533","article-title":"AntMan: Dynamic scaling on GPU clusters for deep learning","volume-title":"Proc. 14th USENIX Symp. Operating Syst. Des. Implementation","author":"Xiao"},{"key":"ref11","volume-title":"Kubernetes: Up and Running: Dive into the Future of Infrastructure","author":"Hightower","year":"2017"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1145\/2523616.2523633"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1145\/2287076.2287091"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/IPDPS.2016.73"},{"key":"ref15","article-title":"Characterization and prediction of performance interference on mediated passthrough GPUs for interference-aware scheduler","volume-title":"Proc. 11th USENIX Conf. Hot Topics Cloud Comput.","author":"Xu"},{"key":"ref16","article-title":"TimeGraph: GPU scheduling for real-time multi-tasking environments","volume-title":"Proc. USENIX Conf. Usenix Annu. Tech. Conf.","author":"Kato"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/CLUSTER.2019.8891040"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1145\/3369583.3392679"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/IISWC47752.2019.9042047"},{"key":"ref20","first-page":"30","article-title":"A systematic methodology for analysis of deep learning hardware and software platforms","volume-title":"Proc. 3rd Mach. Learn. Syst.","author":"Wang"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1145\/3037697.3037700"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1145\/2954679.2872368"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-60239-0_33"},{"key":"ref24","first-page":"6105","article-title":"EfficientNet: Rethinking model scaling for convolutional neural networks","author":"Tan","year":"2019"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.5555\/3291168.3291211"},{"key":"ref26","first-page":"485","article-title":"Tiresias: A GPU cluster manager for distributed deep learning","volume-title":"Proc. 16th USENIX Conf. Netw. Syst. Des. Implementation","author":"Gu"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/L-CA.2009.4"},{"key":"ref28","first-page":"248","article-title":"Bubble-up: Increasing utilization in modern warehouse scale computers via sensible co-locations","volume-title":"Proc. 44th Annu. IEEE\/ACM Int. Symp. Microarchit.","author":"Mars"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1145\/2499368.2451125"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1145\/2644865.2541941"},{"key":"ref31","first-page":"219","article-title":"DeepDive: Transparently identifying and managing performance interference in virtualized environments","volume-title":"Proc. USENIX Conf. Annu. Tech. Conf.","author":"Novakovic"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1145\/2576779.2576780"},{"key":"ref33","article-title":"Learning multiple layers of features from tiny images","volume-title":"Citeseer","author":"Krizhevsky","year":"2009"},{"key":"ref34","article-title":"Pointer sentinel mixture models","author":"Merity","year":"2016"},{"key":"ref35","article-title":"Shared task: Machine translation of news"},{"key":"ref36","article-title":"MobileNets: Efficient convolutional neural networks for mobile vision applications","author":"Howard","year":"2017"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00474"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00140"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.4324\/9781410605337-29"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref41","article-title":"Very deep convolutional networks for large-scale image recognition","author":"Simonyan","year":"2014","journal-title":"Proc. Int. Conf. Learn. Representations"},{"key":"ref42","article-title":"SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and $<$< 0.5 MB model size","author":"Iandola","year":"2016"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.243"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01264-9_8"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00293"},{"key":"ref46","first-page":"4470","article-title":"Dual path networks","volume-title":"Proc. 31st Int. Conf. Neural Inf. Process. Syst.","author":"Chen"},{"key":"ref47","article-title":"ProxylessNAS: Direct neural architecture search on target task and hardware","author":"Cai","year":"2018"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.668"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.634"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1049\/cp:19991218"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.3115\/v1\/D14-1179"},{"key":"ref52","first-page":"8024","article-title":"Pytorch: An imperative style, high-performance deep learning library","volume-title":"Proc. Int. Conf. Neural Inf. Process. Syst.","author":"Paszke"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1145\/3190508.3190517"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2021.3062721"},{"key":"ref55","first-page":"3149","article-title":"LightGBM: A highly efficient gradient boosting decision tree","volume-title":"Proc. 31st Int. Conf. Neural Inf. Process. Syst.","author":"Ke"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1145\/2939672.2939785"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1177\/154405910408300516"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1145\/2485922.2485975"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1109\/DAC18072.2020.9218702"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1109\/90.298432"},{"key":"ref61","first-page":"82","article-title":"PLink: Discovering and exploiting locality for accelerated distributed training on the public cloud","volume-title":"Proc. 3rd Mach. Learn. Syst.","author":"Luo"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2020.2970013"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1109\/SBAC-PAD.2019.00035"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.14778\/3415478.3415539"},{"key":"ref65","doi-asserted-by":"publisher","DOI":"10.1145\/3419111.3421307"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.1145\/3341301.3359642"},{"key":"ref67","article-title":"Paleo: A performance model for deep neural networks","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Qi"}],"container-title":["IEEE Transactions on Parallel and Distributed Systems"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/71\/9465725\/09428512.pdf?arnumber=9428512","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,25]],"date-time":"2026-10-25T04:01:50Z","timestamp":1792900910000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9428512\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,1,1]]},"references-count":67,"journal-issue":{"issue":"1"},"URL":"https:\/\/doi.org\/10.1109\/tpds.2021.3079202","relation":{},"ISSN":["1045-9219","1558-2183","2161-9883"],"issn-type":[{"value":"1045-9219","type":"print"},{"value":"1558-2183","type":"electronic"},{"value":"2161-9883","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,1,1]]}}}