{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,6,18]],"date-time":"2025-06-18T04:23:54Z","timestamp":1750220634835,"version":"3.41.0"},"reference-count":40,"publisher":"Association for Computing Machinery (ACM)","issue":"2","license":[{"start":{"date-parts":[[2020,10,23]],"date-time":"2020-10-23T00:00:00Z","timestamp":1603411200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"funder":[{"DOI":"10.13039\/100006132","name":"U.S. Department of Energy Office of Science","doi-asserted-by":"crossref","award":["DE-AC02-06CH11357"],"award-info":[{"award-number":["DE-AC02-06CH11357"]}],"id":[{"id":"10.13039\/100006132","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Des. Autom. Electron. Syst."],"published-print":{"date-parts":[[2021,3,31]]},"abstract":"<jats:p>The integration of FPGAs into large-scale computing systems is gaining attention. In these systems, real-time data handling for networking, tasks for scientific computing, and machine learning can be executed with customized datapaths on reconfigurable fabric within heterogeneous compute nodes. At the same time, thermal management, particularly battling the cooling cost and guaranteeing the reliability, is a continuing concern. The introduction of new heterogeneous components into HPC nodes only adds further complexities to thermal modeling and management. The thermal behavior of multi-FPGA systems deployed within large compute clusters is less explored. In this article, we first show that the thermal behaviors of different FPGAs of the same generation can vary due to their physical locations in a rack and process variation, even though they are running the same tasks. We present a machine learning\u2013based model to capture the thermal behavior of each individual FPGA in the cluster. We then propose two thermal management strategies guided by our thermal model. First, we mitigate thermal variation and hotspots across the cluster by proactive thermal-aware task placement. Under the tested system and benchmarks, we achieve up to 26.4\u00b0 C and on average 13.3\u00b0 C system temperature reduction with no performance penalty. Second, we utilize this thermal model to guide HLS parameter tuning at the task design stage to achieve improved thermal response after deployment.<\/jats:p>","DOI":"10.1145\/3423494","type":"journal-article","created":{"date-parts":[[2020,10,23]],"date-time":"2020-10-23T21:09:36Z","timestamp":1603487376000},"page":"1-17","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":3,"title":["Thermal Management for FPGA Nodes in HPC Systems"],"prefix":"10.1145","volume":"26","author":[{"given":"Yingyi","family":"Luo","sequence":"first","affiliation":[{"name":"Northwestern University, Evanston, Illinois, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Joshua C.","family":"Zhao","sequence":"additional","affiliation":[{"name":"Northwestern University, Evanston, Illinois, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Arnav","family":"Aggarwal","sequence":"additional","affiliation":[{"name":"William Fremd High School, Palatine, Illinois, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Seda","family":"Ogrenci-Memik","sequence":"additional","affiliation":[{"name":"Northwestern University, Evanston, Illinois, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Kazutomo","family":"Yoshii","sequence":"additional","affiliation":[{"name":"Argonne National Laboratory, Argonne, Illinois, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"320","published-online":{"date-parts":[[2020,10,23]]},"reference":[{"key":"e_1_2_1_1_1","doi-asserted-by":"publisher","DOI":"10.1145\/2617596"},{"key":"e_1_2_1_2_1","unstructured":"AWS Amazon. 2017. Amazon EC2 F1 Instances. Retrieved from https:\/\/aws.amazon.com\/ec2\/instance-types\/f1.  AWS Amazon. 2017. Amazon EC2 F1 Instances. Retrieved from https:\/\/aws.amazon.com\/ec2\/instance-types\/f1."},{"volume-title":"Joseph James Gebis, Parry Husbands, Kurt Keutzer, David A. Patterson, William Lester Plishker, John Shalf, Samuel Webb Williams, et\u00a0al.","year":"2006","author":"Asanovic Krste","key":"e_1_2_1_3_1"},{"key":"e_1_2_1_4_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.future.2018.11.028"},{"key":"e_1_2_1_5_1","doi-asserted-by":"publisher","DOI":"10.5555\/3195638.3195647"},{"key":"e_1_2_1_6_1","doi-asserted-by":"publisher","DOI":"10.1109\/EUC.2015.18"},{"volume-title":"Reliability Report","author":"Intel Corporation","key":"e_1_2_1_7_1"},{"key":"e_1_2_1_8_1","unstructured":"Intel Corporation. 2019. Intel FPGA SDK for OpenCL Design Examples. Retrieved from www.intel.com\/content\/www\/us\/en\/programmable\/products\/design-software\/embedded-software-developers\/opencl\/support.html#design-examples.  Intel Corporation. 2019. Intel FPGA SDK for OpenCL Design Examples. Retrieved from www.intel.com\/content\/www\/us\/en\/programmable\/products\/design-software\/embedded-software-developers\/opencl\/support.html#design-examples."},{"volume-title":"Pecht","year":"2014","author":"Dai Jun","key":"e_1_2_1_9_1"},{"key":"e_1_2_1_10_1","unstructured":"Azure DevOps. 2018. VS Marketplace outage. Retrieved from https:\/\/devblogs.microsoft.com\/devopsservice\/?p=17535.  Azure DevOps. 2018. VS Marketplace outage. Retrieved from https:\/\/devblogs.microsoft.com\/devopsservice\/?p=17535."},{"key":"e_1_2_1_11_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.rser.2013.12.007"},{"volume-title":"Intel Arria 10 Device Datasheet","author":"Intel Corporation 2019.","key":"e_1_2_1_12_1"},{"volume-title":"Intel Quartus Prime Pro Edition User Guide: Power Analysis and Optimization","author":"Intel Corporation 2020.","key":"e_1_2_1_13_1"},{"key":"e_1_2_1_14_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.micpro.2013.05.001"},{"key":"e_1_2_1_15_1","doi-asserted-by":"publisher","DOI":"10.1145\/2567658"},{"key":"e_1_2_1_16_1","doi-asserted-by":"publisher","DOI":"10.1145\/1562514.1562518"},{"key":"e_1_2_1_17_1","doi-asserted-by":"publisher","DOI":"10.1109\/TVLSI.2017.2707120"},{"key":"e_1_2_1_18_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCD.2018.00086"},{"key":"e_1_2_1_19_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCCRI.2015.10"},{"key":"e_1_2_1_20_1","doi-asserted-by":"publisher","DOI":"10.5555\/1509456.1509613"},{"key":"e_1_2_1_21_1","doi-asserted-by":"publisher","DOI":"10.1145\/3180263"},{"key":"e_1_2_1_22_1","doi-asserted-by":"publisher","DOI":"10.1063\/1.1699114"},{"key":"e_1_2_1_23_1","doi-asserted-by":"publisher","DOI":"10.1109\/ISCAS.2006.1693601"},{"volume-title":"Proceedings of the European Regional Science Association Congress (ERSA). 56--62","year":"2006","author":"Mukherjee Rajarshi","key":"e_1_2_1_24_1"},{"volume-title":"Proceedings of the IEEE\/ACM International Conference on Computer-aided Design (ICCAD\u201906)","year":"2006","author":"Mukherjee Rajarshi","key":"e_1_2_1_25_1"},{"key":"e_1_2_1_26_1","doi-asserted-by":"publisher","DOI":"10.1145\/1950413.1950437"},{"key":"e_1_2_1_27_1","first-page":"2825","article-title":"Scikit-learn: Machine learning in Python","author":"Pedregosa Fabian","year":"2011","journal-title":"J. Mach. Learn. Res. 12"},{"volume-title":"Proceedings of the Workshop on Energy-efficient Design.","author":"Pelley Steven","key":"e_1_2_1_28_1"},{"key":"e_1_2_1_29_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCCN.2011.6005802"},{"key":"e_1_2_1_30_1","doi-asserted-by":"publisher","DOI":"10.1109\/ISVLSI.2007.11"},{"volume-title":"Proceedings of the USENIX Conference on Operational Machine Learning (OpML\u201919)","year":"2019","author":"Soifer Jonathan","key":"e_1_2_1_31_1"},{"volume-title":"Dynamic power variations in data centers and network rooms. Amer. Pow. Convers. White Paper 5","year":"2005","author":"Spitaels Jim","key":"e_1_2_1_32_1"},{"volume-title":"Proceedings of the 18th Annual ACM\/SIGDA International Symposium on Field Programmable Gate Arrays (FPGA\u201910)","author":"Stott Edward A.","key":"e_1_2_1_33_1"},{"key":"e_1_2_1_34_1","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2008.111"},{"key":"e_1_2_1_35_1","doi-asserted-by":"publisher","DOI":"10.1145\/3061639.3062251"},{"key":"e_1_2_1_36_1","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA.2016.7446058"},{"key":"e_1_2_1_37_1","unstructured":"Xilinx 2019. Vivado Design Suite User Guide: Power Analysis and Optimization. Xilinx. Retrieved from https:\/\/www.xilinx.com\/support\/documentation\/sw_manuals\/xilinx2019_1\/ug907-vivado-power-analysis-optimization.pdf.  Xilinx 2019. Vivado Design Suite User Guide: Power Analysis and Optimization. Xilinx. Retrieved from https:\/\/www.xilinx.com\/support\/documentation\/sw_manuals\/xilinx2019_1\/ug907-vivado-power-analysis-optimization.pdf."},{"key":"e_1_2_1_38_1","article-title":"Machine learning-based temperature prediction for runtime thermal management across system components","volume":"29","author":"Zhang Kaicheng","year":"2017","journal-title":"IEEE Trans. Parallel Distrib. Syst."},{"key":"e_1_2_1_39_1","doi-asserted-by":"publisher","DOI":"10.1109\/IPDPS.2015.37"},{"key":"e_1_2_1_40_1","doi-asserted-by":"publisher","DOI":"10.5555\/3014904.3014951"}],"container-title":["ACM Transactions on Design Automation of Electronic Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3423494","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3423494","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,17]],"date-time":"2025-06-17T22:01:50Z","timestamp":1750197710000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3423494"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,10,23]]},"references-count":40,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2021,3,31]]}},"alternative-id":["10.1145\/3423494"],"URL":"https:\/\/doi.org\/10.1145\/3423494","relation":{},"ISSN":["1084-4309","1557-7309"],"issn-type":[{"type":"print","value":"1084-4309"},{"type":"electronic","value":"1557-7309"}],"subject":[],"published":{"date-parts":[[2020,10,23]]},"assertion":[{"value":"2020-05-01","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2020-09-01","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2020-10-23","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}