{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,8]],"date-time":"2025-10-08T15:57:35Z","timestamp":1759939055634,"version":"3.41.2"},"publisher-location":"New York, NY, USA","reference-count":39,"publisher":"ACM","funder":[{"name":"NSF","award":["CNS-1730158, ACI-1540112, ACI-1541349, OAC-1826967, OPP-2042807, OAC-2112167, CNS-2100237, CNS-2120019"],"award-info":[{"award-number":["CNS-1730158, ACI-1540112, ACI-1541349, OAC-1826967, OPP-2042807, OAC-2112167, CNS-2100237, CNS-2120019"]}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2025,7,20]]},"DOI":"10.1145\/3708035.3736049","type":"proceedings-article","created":{"date-parts":[[2025,7,18]],"date-time":"2025-07-18T12:10:41Z","timestamp":1752840641000},"page":"1-5","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["SuperSONIC: Cloud-Native Infrastructure for ML Inferencing"],"prefix":"10.1145","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-7874-2480","authenticated-orcid":false,"given":"Dmitry","family":"Kondratyev","sequence":"first","affiliation":[{"name":"Purdue University, West Lafayette, IN, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9524-8943","authenticated-orcid":false,"given":"Benedikt","family":"Riedel","sequence":"additional","affiliation":[{"name":"University of Wisconsin-Madison, Madison, WI, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2204-5731","authenticated-orcid":false,"given":"Yuan-Tang","family":"Chou","sequence":"additional","affiliation":[{"name":"University of Washington, Seattle, WA, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1020-1108","authenticated-orcid":false,"given":"Miles","family":"Cochran-Branson","sequence":"additional","affiliation":[{"name":"University of Washington, Seattle, WA, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1225-537X","authenticated-orcid":false,"given":"Noah","family":"Paladino","sequence":"additional","affiliation":[{"name":"Massachusetts Institute of Technology, Cambridge, MA, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7841-3919","authenticated-orcid":false,"given":"David","family":"Schultz","sequence":"additional","affiliation":[{"name":"University of Wisconsin-Madison, Madison, WI, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9012-395X","authenticated-orcid":false,"given":"Mia","family":"Liu","sequence":"additional","affiliation":[{"name":"Purdue University, West Lafayette, IN, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5076-7096","authenticated-orcid":false,"given":"Javier","family":"Duarte","sequence":"additional","affiliation":[{"name":"University of California San Diego, San Diego, CA, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8189-3741","authenticated-orcid":false,"given":"Philip","family":"Harris","sequence":"additional","affiliation":[{"name":"Massachusetts Institute of Technology, Cambridge, MA, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6214-8500","authenticated-orcid":false,"given":"Shih-Chieh","family":"Hsu","sequence":"additional","affiliation":[{"name":"University of Washington, Seattle, WA, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"320","published-online":{"date-parts":[[2025,7,18]]},"reference":[{"key":"e_1_3_3_1_2_2","unstructured":"T. Akutsu et\u00a0al. 2020. Overview of KAGRA: Detector design and construction history. arxiv:https:\/\/arXiv.org\/abs\/2005.05574\u00a0[physics.ins-det] https:\/\/arxiv.org\/abs\/2005.05574"},{"key":"e_1_3_3_1_3_2","unstructured":"Anton Shilov. 2024. Elon Musk plans to scale the xAI supercomputer to a million GPUs \u2014 currently at over 100 000 H100 GPUs and counting. https:\/\/www.tomshardware.com\/tech-industry\/artificial-intelligence\/elon-musk-plans-to-scale-the-xai-supercomputer-to-a-million-gpus-currently-at-over-100-000-h100-gpus-and-counting Accessed: 2025-03-18."},{"key":"e_1_3_3_1_4_2","unstructured":"Ben Wodecki. 2024. Musk to double xAI\u2019s Colossus cluster to 200K GPUs for Grok 3 training. https:\/\/www.capacitymedia.com\/article\/2dydrju46zsuc1mnwu800\/news\/article-musk-to-double-xais-colossus-cluster-to-200k-gpus-for-grok-3-training Accessed: 2025-03-18."},{"key":"e_1_3_3_1_5_2","unstructured":"CERN. 2025. High-Luminosity LHC. https:\/\/home.cern\/science\/accelerators\/high-luminosity-lhc Accessed: 2025-03-18."},{"key":"e_1_3_3_1_6_2","unstructured":"Cloud Native Computing Foundation. 2015. CNCF: Graduated and Incubating Projects. https:\/\/www.cncf.io\/projects\/. Accessed: 2025-03-18."},{"key":"e_1_3_3_1_7_2","doi-asserted-by":"publisher","unstructured":"Javier Duarte et\u00a0al. 2019. FPGA-accelerated machine learning inference as a service for particle physics computing. Comput. Softw. Big Sci. 3 1 (2019) 13. 10.1007\/s41781-019-0027-2 arxiv:https:\/\/arXiv.org\/abs\/1904.08986\u00a0[physics.data-an]","DOI":"10.1007\/s41781-019-0027-2"},{"key":"e_1_3_3_1_8_2","unstructured":"Javier\u00a0M. Duarte. 2024. Novel machine learning applications at the LHC. arxiv:https:\/\/arXiv.org\/abs\/2409.20413\u00a0[hep-ex] https:\/\/arxiv.org\/abs\/2409.20413"},{"key":"e_1_3_3_1_9_2","unstructured":"GitHub. 2025. GitHub Actions documentation. https:\/\/docs.github.com\/en\/actions Accessed: 2025-03-20."},{"key":"e_1_3_3_1_10_2","unstructured":"Grafana Labs. 2014. Grafana: Open Source Analytics & Monitoring. https:\/\/grafana.com. Accessed: 2025-03-18."},{"key":"e_1_3_3_1_11_2","unstructured":"Grafana Labs. 2020. Grafana Tempo: Distributed Tracing for Cloud-Native Systems. https:\/\/grafana.com\/oss\/tempo. Accessed: 2025-03-18."},{"key":"e_1_3_3_1_12_2","unstructured":"Jiacheng Guo and Santosh Bhavani. 2021. Reduce ML inference costs on Amazon SageMaker with hardware and software acceleration. https:\/\/aws.amazon.com\/blogs\/machine-learning\/reduce-ml-inference-costs-on-amazon-sagemaker-with-hardware-and-software-acceleration\/ Accessed: 2025-03-18."},{"key":"e_1_3_3_1_13_2","doi-asserted-by":"publisher","unstructured":"Aram Hayrapetyan et\u00a0al. 2024. Portable Acceleration of CMS Computing Workflows with Coprocessors as a Service. Comput. Softw. Big Sci. 8 1 (2024) 17. 10.1007\/s41781-024-00124-1 arxiv:https:\/\/arXiv.org\/abs\/2402.15366\u00a0[physics.ins-det]","DOI":"10.1007\/s41781-024-00124-1"},{"key":"e_1_3_3_1_14_2","unstructured":"Aya Ishihara. 2019. The IceCube Upgrade \u2013 Design and Science Goals. arxiv:https:\/\/arXiv.org\/abs\/1908.09441\u00a0[astro-ph.HE] https:\/\/arxiv.org\/abs\/1908.09441"},{"key":"e_1_3_3_1_15_2","unstructured":"Kim Martineau. 2023. What is AI inferencing?https:\/\/research.ibm.com\/blog\/AI-inference-explained Accessed: 2025-03-18."},{"key":"e_1_3_3_1_16_2","doi-asserted-by":"publisher","DOI":"10.5281\/zenodo.14816533"},{"key":"e_1_3_3_1_17_2","doi-asserted-by":"publisher","unstructured":"Jeffrey Krupa et\u00a0al. 2021. GPU coprocessors as a service for deep learning inference in high energy physics. Mach. Learn. Sci. Tech. 2 3 (2021) 035005. 10.1088\/2632-2153\/abec21 arxiv:https:\/\/arXiv.org\/abs\/2007.10359\u00a0[physics.comp-ph]","DOI":"10.1088\/2632-2153\/abec21"},{"key":"e_1_3_3_1_18_2","unstructured":"Nobel Prize Committee. 2024. 2024 Nobel Prizes. https:\/\/www.nobelprize.org\/all-nobel-prizes-2024\/ Accessed: 2025-03-18."},{"key":"e_1_3_3_1_19_2","volume-title":"PyTriton: Framework facilitating NVIDIA Triton Inference Server usage in Python environments","author":"Corporation NVIDIA","unstructured":"NVIDIA Corporation. [n. d.]. PyTriton: Framework facilitating NVIDIA Triton Inference Server usage in Python environments. https:\/\/github.com\/triton-inference-server\/pytriton"},{"key":"e_1_3_3_1_20_2","volume-title":"Triton Inference Server: An Optimized Cloud and Edge Inferencing Solution.","author":"Corporation NVIDIA","year":"2016","unstructured":"NVIDIA Corporation. 2016. Triton Inference Server: An Optimized Cloud and Edge Inferencing Solution.https:\/\/github.com\/triton-inference-server\/server Accessed: 2025-03-18."},{"key":"e_1_3_3_1_21_2","volume-title":"Triton Performance Analyzer","author":"Corporation NVIDIA","year":"2021","unstructured":"NVIDIA Corporation. 2021. Triton Performance Analyzer. https:\/\/github.com\/triton-inference-server\/perf_analyzer\/ Accessed: 2025-06-18."},{"key":"e_1_3_3_1_22_2","unstructured":"OpenAI. 2024. ChatGPT: An AI language model by OpenAI. https:\/\/chat.openai.com. Accessed: 2025-03-26."},{"key":"e_1_3_3_1_23_2","doi-asserted-by":"publisher","unstructured":"Huilin Qu and Loukas Gouskos. 2020. ParticleNet: Jet Tagging via Particle Clouds. Phys. Rev. D 101 (2020) 056019. 10.1103\/PhysRevD.101.056019 arxiv:https:\/\/arXiv.org\/abs\/1902.08570\u00a0[hep-ph]","DOI":"10.1103\/PhysRevD.101.056019"},{"key":"e_1_3_3_1_24_2","doi-asserted-by":"publisher","unstructured":"Dylan\u00a0Sheldon Rankin et\u00a0al. 2020. FPGAs-as-a-Service Toolkit (FaaST). 10.1109\/H2RC51942.2020.00010 arxiv:https:\/\/arXiv.org\/abs\/2010.08556\u00a0[physics.comp-ph]","DOI":"10.1109\/H2RC51942.2020.00010"},{"key":"e_1_3_3_1_25_2","doi-asserted-by":"publisher","unstructured":"Rind Ofer Benjamin Douglas Bryant Lincoln Caramarcu Costin Gardner Robert Golnaraghi Farnaz Hollowell Christopher Hu Fengping Jordan David Stephen Judith Vukotic Ilija and Yang Wei. 2024. The Creation and Evolution of the US ATLAS Shared Analysis Facilities. EPJ Web of Conf. 295 (2024) 07043. 10.1051\/epjconf\/202429507043","DOI":"10.1051\/epjconf\/202429507043"},{"key":"e_1_3_3_1_26_2","doi-asserted-by":"publisher","DOI":"10.1145\/3219104.3219108"},{"key":"e_1_3_3_1_27_2","doi-asserted-by":"publisher","DOI":"10.1109\/SuperCompCloud51944.2020.00011"},{"key":"e_1_3_3_1_28_2","doi-asserted-by":"publisher","DOI":"10.1145\/3491418.3530766"},{"key":"e_1_3_3_1_29_2","unstructured":"The Envoy Proxy Authors. 2016. Envoy Proxy: A Cloud-Native High-Performance Edge\/Middle\/Service Proxy. https:\/\/www.envoyproxy.io. Accessed: 2025-03-18."},{"key":"e_1_3_3_1_30_2","unstructured":"The Helm Authors. 2016. Helm: The Package Manager for Kubernetes. https:\/\/helm.sh. Accessed: 2025-03-18."},{"key":"e_1_3_3_1_31_2","unstructured":"The KEDA Authors. 2018. KEDA: Kubernetes Event-Driven Autoscaling. https:\/\/keda.sh. Accessed: 2025-03-18."},{"key":"e_1_3_3_1_32_2","unstructured":"The KServe Authors. 2015. KServe Docs. https:\/\/kserve.github.io\/website\/latest\/. Accessed: 2025-03-18."},{"key":"e_1_3_3_1_33_2","unstructured":"The Kubernetes Authors. 2014. Kubernetes: Production-Grade Container Orchestration. https:\/\/kubernetes.io Accessed: 2025-03-18."},{"key":"e_1_3_3_1_34_2","unstructured":"The Kubernetes Authors. 2018. kind: Kubernetes in Docker. https:\/\/kind.sigs.k8s.io\/ Accessed: 2025-03-20."},{"key":"e_1_3_3_1_35_2","unstructured":"The OpenTelemetry Authors. 2019. OpenTelemetry: Observability for Cloud-Native Software. https:\/\/opentelemetry.io. Accessed: 2025-03-18."},{"key":"e_1_3_3_1_36_2","unstructured":"The Prometheus Authors. 2012. Prometheus: Monitoring System & Time Series Database. https:\/\/prometheus.io. Accessed: 2025-03-18."},{"key":"e_1_3_3_1_37_2","unstructured":"The vLLM Authors. 2015. vLLM Docs. https:\/\/docs.vllm.ai\/en\/latest\/. Accessed: 2025-03-18"},{"key":"e_1_3_3_1_38_2","doi-asserted-by":"publisher","unstructured":"Michael Wang Tingjun Yang Maria Acosta\u00a0Flechas Philip Harris Benjamin Hawks Burt Holzman Kyle Knoepfel Jeffrey Krupa Kevin Pedro and Nhan Tran. 2021. GPU-Accelerated Machine Learning Inference as a Service for Computing in Neutrino Experiments. Front. Big Data 3 (2021) 604083. 10.3389\/fdata.2020.604083 arxiv:https:\/\/arXiv.org\/abs\/2009.04509\u00a0[physics.comp-ph]","DOI":"10.3389\/fdata.2020.604083"},{"key":"e_1_3_3_1_39_2","unstructured":"Frank W\u00fcrthwein et\u00a0al. 2025. National Research Platform. https:\/\/nationalresearchplatform.org\/. [Accessed 20-03-2025]."},{"key":"e_1_3_3_1_40_2","doi-asserted-by":"publisher","unstructured":"Haoran Zhao Yuan-Tang Chou Yao Yao Xiangyang Ju Yongbin Feng William\u00a0Patrick McCormack Miles Cochran-Branson Jan-Frederik Schulte Miaoyuan Liu Javier Duarte Philip Harris Shih-Chieh Hsu Kevin Pedro and Nhan Tran. 2025. Track reconstruction as a service for collider physics. Journal of Instrumentation 20 06 (jun 2025) P06002. 10.1088\/1748-0221\/20\/06\/P06002","DOI":"10.1088\/1748-0221\/20\/06\/P06002"}],"event":{"name":"PEARC '25: Practice and Experience in Advanced Research Computing","location":"Columbus Ohio USA","acronym":"PEARC '25","sponsor":["SIGAPP ACM Special Interest Group on Applied Computing","SIGHPC ACM Special Interest Group on High Performance Computing, Special Interest Group on High Performance Computing"]},"container-title":["Practice and Experience in Advanced Research Computing 2025: The Power of Collaboration"],"original-title":[],"link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3708035.3736049","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,18]],"date-time":"2025-07-18T12:32:20Z","timestamp":1752841940000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3708035.3736049"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7,18]]},"references-count":39,"alternative-id":["10.1145\/3708035.3736049","10.1145\/3708035"],"URL":"https:\/\/doi.org\/10.1145\/3708035.3736049","relation":{},"subject":[],"published":{"date-parts":[[2025,7,18]]},"assertion":[{"value":"2025-07-18","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}