{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,24]],"date-time":"2026-06-24T18:58:09Z","timestamp":1782327489073,"version":"3.54.5"},"reference-count":75,"publisher":"Association for Computing Machinery (ACM)","issue":"3","license":[{"start":{"date-parts":[[2024,5,29]],"date-time":"2024-05-29T00:00:00Z","timestamp":1716940800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"funder":[{"DOI":"10.13039\/501100006374","name":"NSF","doi-asserted-by":"publisher","award":["2337806"],"award-info":[{"award-number":["2337806"]}],"id":[{"id":"10.13039\/501100006374","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Proc. ACM Manag. Data"],"published-print":{"date-parts":[[2024,5,29]]},"abstract":"<jats:p>Cloud functions, exemplified by AWS Lambda and Azure Functions, are emerging as a new computing paradigm in the cloud. They provide elastic, serverless, and low-cost cloud computing, making them highly suitable for bursty and sparse workloads, which are quite common in practice. Thus, there is a new trend in designing data systems that leverage cloud functions. In this paper, we focus on vector databases, which have recently gained significant attention partly due to large language models. In particular, we investigate how to use cloud functions to build high-performance and cost-efficient vector databases. This presents significant challenges in terms of how to perform sharding, how to reduce communication overhead, and how to minimize cold-start times.<\/jats:p>\n          <jats:p>In this paper, we introduce Vexless, the first vector database system optimized for cloud functions. We present three optimizations to address the challenges. To perform sharding, we propose a global coordinator (orchestrator) that assigns workloads to Cloud function instances based on their available hardware resources. To overcome communication overhead, we propose the use of stateful cloud functions, eliminating the need for costly communications during synchronization. To minimize cold-start overhead, we introduce a workload-aware Cloud function lifetime management strategy. Vexless has been implemented using Azure Functions. Experimental results demonstrate that Vexless can significantly reduce costs, especially on bursty and sparse workloads, compared to cloud VM instances, while achieving similar or higher query performance and accuracy.<\/jats:p>","DOI":"10.1145\/3654990","type":"journal-article","created":{"date-parts":[[2024,5,30]],"date-time":"2024-05-30T09:44:53Z","timestamp":1717062293000},"page":"1-26","source":"Crossref","is-referenced-by-count":25,"title":["Vexless: A Serverless Vector Data Management System Using Cloud Functions"],"prefix":"10.1145","volume":"2","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-9297-5902","authenticated-orcid":false,"given":"Yongye","family":"Su","sequence":"first","affiliation":[{"name":"Purdue University, West Lafayette, IN, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-3783-5472","authenticated-orcid":false,"given":"Yinqi","family":"Sun","sequence":"additional","affiliation":[{"name":"Purdue University, West Lafayette, IN, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8165-166X","authenticated-orcid":false,"given":"Minjia","family":"Zhang","sequence":"additional","affiliation":[{"name":"University of Illinois Urbana-Champaign, Urbana, IL, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3039-1175","authenticated-orcid":false,"given":"Jianguo","family":"Wang","sequence":"additional","affiliation":[{"name":"Purdue University, West Lafayette, IN, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2024,5,30]]},"reference":[{"key":"e_1_2_1_1_1","unstructured":"[n. d.]. Alibaba Cloud: Manage Stateful Asynchronous Invocations. https:\/\/www.alibabacloud.com\/help\/en\/fc\/developer-reference\/manage-stateful-asynchronous-invocations."},{"key":"e_1_2_1_2_1","unstructured":"[n. d.]. Alibaba Cloud: Message Service (MNS). https:\/\/www.alibabacloud.com\/product\/message-service."},{"key":"e_1_2_1_3_1","unstructured":"[n. d.]. Amazon Simple Queue Service. https:\/\/aws.amazon.com\/sqs."},{"key":"e_1_2_1_4_1","unstructured":"[n. d.]. AWS Lambda - Serverless Compute - Amazon Web Services. https:\/\/aws.amazon.com\/lambda."},{"key":"e_1_2_1_5_1","unstructured":"[n. d.]. AWS Step Functions. https:\/\/aws.amazon.com\/step-functions."},{"key":"e_1_2_1_6_1","unstructured":"[n. d.]. Azure Functions - Serverless Code. https:\/\/azure.microsoft.com\/services\/functions."},{"key":"e_1_2_1_7_1","unstructured":"[n. d.]. Azure Functions Scale and Hosting. https:\/\/learn.microsoft.com\/azure\/azure-functions\/functions-scale."},{"key":"e_1_2_1_8_1","unstructured":"[n. d.]. Benchmarks for Billion-Scale Similarity Search. https:\/\/research.yandex.com\/blog\/benchmarks-for-billion-scale-similarity-search."},{"key":"e_1_2_1_9_1","unstructured":"[n. d.]. Cloud Functions: Serverless Computing Google Cloud. https:\/\/cloud.google.com\/functions."},{"key":"e_1_2_1_10_1","unstructured":"[n. d.]. Cold Starts in Azure Functions. https:\/\/mikhail.io\/serverless\/coldstarts\/azure."},{"key":"e_1_2_1_11_1","unstructured":"[n. d.]. Compute Optimized F Series - Azure Virtual Machines. https:\/\/learn.microsoft.com\/en-us\/azure\/virtual-machines\/sizes-compute."},{"key":"e_1_2_1_12_1","unstructured":"[n. d.]. Google Cloud Pub\/Sub. https:\/\/cloud.google.com\/pubsub."},{"key":"e_1_2_1_13_1","unstructured":"[n. d.]. Microsoft Azure Durable Functions. https:\/\/learn.microsoft.com\/azure\/azure-functions\/durable\/durable-functions-overview."},{"key":"e_1_2_1_14_1","unstructured":"[n. d.]. Microsoft Azure Queue Storage. https:\/\/learn.microsoft.com\/azure\/storage\/queues\/storage-queues-introduction."},{"key":"e_1_2_1_15_1","unstructured":"[n. d.]. pgvector. https:\/\/github.com\/pgvector\/pgvector."},{"key":"e_1_2_1_16_1","unstructured":"[n. d.]. Pinecone: Vector Database for Vector Search. https:\/\/www.pinecone.io."},{"key":"e_1_2_1_17_1","unstructured":"[n. d.]. Scalability and Performance Targets for Blob storage. https:\/\/learn.microsoft.com\/azure\/storage\/blobs\/scalability-targets."},{"key":"e_1_2_1_18_1","unstructured":"[n. d.]. Vespa (https:\/\/vespa.ai\/)."},{"key":"e_1_2_1_19_1","unstructured":"[n. d.]. What's the \"Average\" Requests Per Second for a Production Web Application? https:\/\/stackoverflow.com\/questions\/373098\/whats-the-average-requests-per-second-for-a-production-web-application."},{"key":"e_1_2_1_20_1","unstructured":"2023. LLM Limitations. (https:\/\/zilliz.com\/use-cases\/llm-retrieval-augmented-generation)."},{"key":"e_1_2_1_21_1","unstructured":"2023. Solving ChatGPT Hallucinations With Vector Embeddings https:\/\/www.youtube.com\/watch?v=FUgp4oaxj-M."},{"key":"e_1_2_1_22_1","volume-title":"Tractable Nonparametric Bayesian Inference in Poisson Processes with Gaussian Process Intensities. In International Conference on Machine Learning (ICML). 9--16","author":"Adams Ryan Prescott","year":"2009","unstructured":"Ryan Prescott Adams, Iain Murray, and David JC MacKay. 2009. Tractable Nonparametric Bayesian Inference in Poisson Processes with Gaussian Process Intensities. In International Conference on Machine Learning (ICML). 9--16."},{"key":"e_1_2_1_23_1","doi-asserted-by":"publisher","DOI":"10.14778\/3352063.3352092"},{"key":"e_1_2_1_24_1","doi-asserted-by":"publisher","DOI":"10.1145\/1327452.1327494"},{"key":"e_1_2_1_25_1","volume-title":"Efficient Indexing of Billion-Scale Datasets of Deep Descriptors. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2055--2063","author":"Babenko Artem","year":"2016","unstructured":"Artem Babenko and Victor Lempitsky. 2016. Efficient Indexing of Billion-Scale Datasets of Deep Descriptors. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2055--2063."},{"key":"e_1_2_1_26_1","doi-asserted-by":"publisher","DOI":"10.1109\/MLSP.2016.7738886"},{"key":"e_1_2_1_27_1","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2006.27"},{"key":"e_1_2_1_28_1","doi-asserted-by":"publisher","DOI":"10.1145\/3589306"},{"key":"e_1_2_1_29_1","first-page":"0","article-title":"Constrained K-means Clustering. Microsoft Research","volume":"20","author":"Bradley Paul S","year":"2000","unstructured":"Paul S Bradley, Kristin P Bennett, and Ayhan Demiriz. 2000. Constrained K-means Clustering. Microsoft Research, Redmond 20, 0 (2000), 0.","journal-title":"Redmond"},{"key":"e_1_2_1_30_1","doi-asserted-by":"publisher","DOI":"10.1016\/S1389-1286(01)00285-7"},{"key":"e_1_2_1_31_1","volume-title":"SPTAG: A Library for Fast Approximate Nearest Neighbor Search. https:\/\/github.com\/Microsoft\/SPTAG.","author":"Chen Qi","year":"2018","unstructured":"Qi Chen, Haidong Wang, Mingqin Li, Gang Ren, Scarlett Li, Jeffery Zhu, Jason Li, Chuanjie Liu, Lintao Zhang, and Jingdong Wang. 2018. SPTAG: A Library for Fast Approximate Nearest Neighbor Search. https:\/\/github.com\/Microsoft\/SPTAG."},{"key":"e_1_2_1_32_1","volume-title":"Proceedings of the International Conference on Neural Information Processing Systems (NeurIPS). 5199--5212","author":"Chen Qi","year":"2021","unstructured":"Qi Chen, Bing Zhao, Haidong Wang, Mingqin Li, Chuanjie Liu, Zengzhong Li, Mao Yang, and Jingdong Wang. 2021. SPANN: Highly-efficient Billion-scale Approximate Nearest Neighborhood Search. In Proceedings of the International Conference on Neural Information Processing Systems (NeurIPS). 5199--5212."},{"key":"e_1_2_1_33_1","volume-title":"Imagenet: A Large-Scale Hierarchical Image Database. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 248--255","author":"Deng Jia","year":"2009","unstructured":"Jia Deng, Wei Dong, Richard Socher, Li-Jia Li, Kai Li, and Li Fei-Fei. 2009. Imagenet: A Large-Scale Hierarchical Image Database. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 248--255."},{"key":"e_1_2_1_34_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2023.109999"},{"key":"e_1_2_1_35_1","doi-asserted-by":"publisher","DOI":"10.1145\/1646396.1646421"},{"key":"e_1_2_1_36_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-023-28390-w"},{"key":"e_1_2_1_37_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.is.2005.01.001"},{"key":"e_1_2_1_38_1","doi-asserted-by":"publisher","DOI":"10.14778\/3303753.3303754"},{"key":"e_1_2_1_39_1","volume-title":"International Conference on Very Large Data Bases (VLDB). 518--529","author":"Gionis Aristides","year":"1999","unstructured":"Aristides Gionis, Piotr Indyk, and Rajeev Motwani. 1999. Similarity Search in High Dimensions via Hashing. In International Conference on Very Large Data Bases (VLDB). 518--529."},{"key":"e_1_2_1_40_1","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2010.57"},{"key":"e_1_2_1_41_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.aeue.2014.04.011"},{"key":"e_1_2_1_42_1","doi-asserted-by":"publisher","DOI":"10.1109\/TBDATA.2019.2921572"},{"key":"e_1_2_1_43_1","volume-title":"Joseph E. Gonzalez, Raluca Ada Popa, Ion Stoica, and David A. Patterson.","author":"Jonas Eric","year":"2019","unstructured":"Eric Jonas, Johann Schleier-Smith, Vikram Sreekanti, Chia-che Tsai, Anurag Khandelwal, Qifan Pu, Vaishaal Shankar, Joao Carreira, Karl Krauth, Neeraja Jayant Yadwadkar, Joseph E. Gonzalez, Raluca Ada Popa, Ion Stoica, and David A. Patterson. 2019. Cloud Programming Simplified: A Berkeley View on Serverless Computing. CoRR abs\/1902.03383 (2019)."},{"key":"e_1_2_1_44_1","volume-title":"Distributed Representations of Sentences and Documents. In International Conference on Machine Learning (ICML). 1188--1196","author":"Quoc","unstructured":"Quoc V. Le and Tomas Mikolov. 2014. Distributed Representations of Sentences and Documents. In International Conference on Machine Learning (ICML). 1188--1196."},{"key":"e_1_2_1_45_1","doi-asserted-by":"publisher","DOI":"10.1145\/3318464.3380600"},{"key":"e_1_2_1_46_1","volume-title":"Cosine Threshold Querying with Optimality Guarantees. In International Conference on Database Theory (ICDT)","volume":"127","author":"Li Yuliang","year":"2019","unstructured":"Yuliang Li, Jianguo Wang, Benjamin S. Pullman, Nuno Bandeira, and Yannis Papakonstantinou. 2019. Index-Based, High-Dimensional, Cosine Threshold Querying with Optimality Guarantees. In International Conference on Database Theory (ICDT), Vol. 127. 11:1--11:20."},{"key":"e_1_2_1_47_1","doi-asserted-by":"publisher","DOI":"10.1145\/3585007"},{"key":"e_1_2_1_48_1","doi-asserted-by":"publisher","DOI":"10.14778\/3397230.3397240"},{"key":"e_1_2_1_49_1","volume-title":"Proceedings of the VLDB Endowment (PVLDB). 950--961","author":"Lv Qin","year":"2007","unstructured":"Qin Lv, William Josephson, Zhe Wang, Moses Charikar, and Kai Li. 2007. Multi-Probe LSH: Efficient Indexing for High-Dimensional Similarity Search. In Proceedings of the VLDB Endowment (PVLDB). 950--961."},{"key":"e_1_2_1_50_1","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2018.2889473"},{"key":"e_1_2_1_51_1","doi-asserted-by":"publisher","DOI":"10.1093\/imamat\/24.1.59"},{"key":"e_1_2_1_52_1","unstructured":"Erik Bernhardsson Martin Aumueller. 2023. ANN-Benchmarks. https:\/\/ann-benchmarks.com."},{"key":"e_1_2_1_53_1","volume-title":"Efficient Estimation of Word Representations in Vector Space. In International Conference on Learning Representations (ICLR).","author":"Mikolov Tomas","year":"2013","unstructured":"Tomas Mikolov, Kai Chen, Greg Corrado, and Jeffrey Dean. 2013. Efficient Estimation of Word Representations in Vector Space. In International Conference on Learning Representations (ICLR)."},{"key":"e_1_2_1_54_1","first-page":"52","article-title":"Cloud Computing for Climate Modelling: Evaluation, Challenges and Benefits","volume":"9","author":"Montes Diego","year":"2020","unstructured":"Diego Montes, Juan A A\u00f1el, David CH Wallom, Peter Uhe, Pablo V Caderno, and Tom\u00e1s F Pena. 2020. Cloud Computing for Climate Modelling: Evaluation, Challenges and Benefits. MDPI Computers 9, 2 (2020), 52.","journal-title":"MDPI Computers"},{"key":"e_1_2_1_55_1","doi-asserted-by":"publisher","DOI":"10.1145\/3318464.3389758"},{"key":"e_1_2_1_56_1","volume-title":"Vector Database Management Techniques and Systems. In Companion of the International Conference on Management of Data (SIGMOD).","author":"Pan James","year":"2024","unstructured":"James Pan, Jianguo Wang, and Guoliang Li. 2024. Vector Database Management Techniques and Systems. In Companion of the International Conference on Management of Data (SIGMOD)."},{"key":"e_1_2_1_57_1","volume-title":"Survey of Vector Database Management Systems. CoRR abs\/2310.14021","author":"Pan James Jie","year":"2023","unstructured":"James Jie Pan, Jianguo Wang, and Guoliang Li. 2023. Survey of Vector Database Management Systems. CoRR abs\/2310.14021 (2023)."},{"key":"e_1_2_1_58_1","doi-asserted-by":"publisher","DOI":"10.1145\/3572848.3577527"},{"key":"e_1_2_1_59_1","doi-asserted-by":"publisher","DOI":"10.1145\/3318464.3380609"},{"key":"e_1_2_1_60_1","volume-title":"Scheduling of Sporadic Tasks with Deadline Constrains in Cloud Environments. In International Conference on Advanced Information Networking and Applications (AINA). 764--771","author":"Pop Florin","year":"2013","unstructured":"Florin Pop, Ciprian Dobre, Valentin Cristea, and Nik Bessis. 2013. Scheduling of Sporadic Tasks with Deadline Constrains in Cloud Environments. In International Conference on Advanced Information Networking and Applications (AINA). 764--771."},{"key":"e_1_2_1_61_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICBAIE52039.2021.9389905"},{"key":"e_1_2_1_62_1","doi-asserted-by":"publisher","DOI":"10.1109\/18.761337"},{"key":"e_1_2_1_63_1","volume-title":"Optimised KD-trees for Fast Image Descriptor Matching. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 1--8.","author":"Silpa-Anan Chanop","unstructured":"Chanop Silpa-Anan and Richard I. Hartley. 2008. Optimised KD-trees for Fast Image Descriptor Matching. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 1--8."},{"key":"e_1_2_1_64_1","doi-asserted-by":"publisher","DOI":"10.1145\/3423211.3425682"},{"key":"e_1_2_1_65_1","volume-title":"Annual Conference on Neural Information Processing Systems (NeurIPS). 13748--13758","author":"Subramanya Suhas Jayaram","year":"2019","unstructured":"Suhas Jayaram Subramanya, Fnu Devvrit, Harsha Vardhan Simhadri, Ravishankar Krishnaswamy, and Rohan Kadekodi. 2019. Rand-NSG: Fast Accurate Billion-point Nearest Neighbor Search on a Single Node. In Annual Conference on Neural Information Processing Systems (NeurIPS). 13748--13758."},{"key":"e_1_2_1_66_1","volume-title":"Going Deeper with Convolutions. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 1--9.","author":"Szegedy Christian","year":"2015","unstructured":"Christian Szegedy, Wei Liu, Yangqing Jia, Pierre Sermanet, Scott Reed, Dragomir Anguelov, Dumitru Erhan, Vincent Vanhoucke, and Andrew Rabinovich. 2015. Going Deeper with Convolutions. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 1--9."},{"key":"e_1_2_1_67_1","volume-title":"USENIX Symposium on Operating Systems Design and Implementation (OSDI). 495--514","author":"Thorpe John","year":"2021","unstructured":"John Thorpe, Yifan Qiao, Jonathan Eyolfson, Shen Teng, Guanzhou Hu, Zhihao Jia, Jinliang Wei, Keval Vora, Ravi Netravali, Miryung Kim, and Guoqing Harry Xu. 2021. Dorylus: Affordable, Scalable, and Accurate GNN Training with Distributed CPU Servers and Serverless Threads. In USENIX Symposium on Operating Systems Design and Implementation (OSDI). 495--514."},{"key":"e_1_2_1_68_1","doi-asserted-by":"publisher","DOI":"10.1145\/3448016.3457550"},{"key":"e_1_2_1_69_1","volume-title":"Peeking Behind the Curtains of Serverless Platforms. In USENIX Annual Technical Conference (USENIX ATC). 133--146","author":"Wang Liang","year":"2018","unstructured":"Liang Wang, Mengyuan Li, Yinqian Zhang, Thomas Ristenpart, and Michael Swift. 2018. Peeking Behind the Curtains of Serverless Platforms. In USENIX Annual Technical Conference (USENIX ATC). 133--146."},{"key":"e_1_2_1_70_1","doi-asserted-by":"publisher","DOI":"10.14778\/3415478.3415541"},{"key":"e_1_2_1_71_1","doi-asserted-by":"publisher","DOI":"10.1145\/3514221.3517905"},{"key":"e_1_2_1_72_1","doi-asserted-by":"publisher","DOI":"10.1109\/IISWC.2011.6114193"},{"key":"e_1_2_1_73_1","volume-title":"Adaptive Function Launching Acceleration in Serverless Computing Platforms. In IEEE International Conference on Parallel and Distributed Systems (ICPADS). 9--16","author":"Xu Zhengjun","year":"2019","unstructured":"Zhengjun Xu, Haitao Zhang, Xin Geng, Qiong Wu, and Huadong Ma. 2019. Adaptive Function Launching Acceleration in Serverless Computing Platforms. In IEEE International Conference on Parallel and Distributed Systems (ICPADS). 9--16."},{"key":"e_1_2_1_74_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICDE60146.2024.00280"},{"key":"e_1_2_1_75_1","doi-asserted-by":"publisher","DOI":"10.1145\/2488388.2488523"}],"container-title":["Proceedings of the ACM on Management of Data"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3654990","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3654990","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,8,21]],"date-time":"2025-08-21T14:41:23Z","timestamp":1755787283000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3654990"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,5,29]]},"references-count":75,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2024,5,29]]}},"alternative-id":["10.1145\/3654990"],"URL":"https:\/\/doi.org\/10.1145\/3654990","relation":{},"ISSN":["2836-6573"],"issn-type":[{"value":"2836-6573","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,5,29]]}}}