Nội dung được dịch bởi AI, chỉ mang tính chất tham khảo
Phân tích độ trễ và tiêu thụ tài nguyên cho phân tích biên không máy chủ
Tóm tắt
Mô hình điện toán không máy chủ, được triển khai bởi các nền tảng Function as a Service (FaaS), có thể mang lại nhiều lợi ích cho việc triển khai các giải pháp phân tích dữ liệu trong môi trường IoT, chẳng hạn như cung cấp tài nguyên linh hoạt và theo yêu cầu, khả năng tự động mở rộng, độ đàn hồi cao, trừu tượng hóa quản lý hạ tầng và mô hình chi phí chi tiết. Tuy nhiên, trong trường hợp các ứng dụng có yêu cầu độ trễ nghiêm ngặt, vấn đề khởi động lạnh trong các nền tảng FaaS có thể đại diện cho một nhược điểm quan trọng. Các kỹ thuật phổ biến nhất để giảm nhẹ vấn đề này, chủ yếu dựa trên việc làm nóng trước các phiên bản và cơ chế tái sử dụng các phiên bản, thường không thích nghi tốt với các hồ sơ ứng dụng khác nhau và, nhìn chung, có thể gây ra chi phí bổ sung về tài nguyên. Trong công trình này, chúng tôi phân tích ảnh hưởng của việc làm nóng trước các phiên bản và tái sử dụng các phiên bản đối với cả độ trễ ứng dụng (thời gian phản hồi) và tiêu thụ tài nguyên, cho một trường hợp sử dụng phân tích dữ liệu điển hình (ứng dụng học máy cho phân loại hình ảnh) với các mẫu dữ liệu đầu vào khác nhau. Hơn nữa, chúng tôi đề xuất mở rộng nền tảng FaaS không máy chủ dựa trên đám mây tập trung cổ điển thành một nền tảng phân tán biên-đám mây hai tầng để đưa nền tảng này gần hơn với nguồn dữ liệu và giảm độ trễ mạng.
Từ khóa
#không máy chủ #phân tích dữ liệu #IoT #khởi động lạnh #FaaS #học máy #độ trễ #tiêu thụ tài nguyênTài liệu tham khảo
Yu T, Wang X (2020) Real-time data analytics in internet of things systems. In: Tian Y, Levy D (eds) Handbook of real-time computing. Springer, Singapore. https://doi.org/10.1007/978-981-4585-87-3_38-1
Atitallah S, Driss M, Boulila W, Ghézala H (2020) Leveraging deep learning and IoT big data analytics to support the smart cities development: review and future directions. Comput Sci Rev 38. https://doi.org/10.1016/j.cosrev.2020.100303
Ellis B (2014) Real-time analytics: techniques to analyze and visualize streaming data. Indianapolis: Wiley; 2014.
Nastic S et al (2017) A serverless real-time data analytics platform for edge computing. IEEE Internet Comput 21(4):64–71. https://doi.org/10.1109/MIC.2017.2911430
López P et al. (2019) ServerMix: tradeoffs and challenges of serverless data analytics. arXiv: 1907.11465v1. doi:https://doi.org/10.48550/arXiv.1907.11465
Castro P, Ishakian V, Muthusamy V, Slominski A (2019) The rise of serverless computing. Comm of the ACM 62(12):44–54. https://doi.org/10.1145/3368454
Jonas E et al. (2019) Cloud programming simplified: a Berkeley view on serverless computing. arXiv:1902.03383v1. https://doi.org/10.48550/arXiv.1902.03383
Baldini I et al (2017) Serverless computing: current trends and open problems. In: Chaudhary S, Somani G, Buyya R (eds) Research advances in cloud computing. Springer, Singapore. https://doi.org/10.1007/978-981-10-5026-8_1
Manner J, Endreß M, Heckel T, Wirtz G (2018) Cold Start Influencing Factors in Function as a Service. In: IEEE/ACM Int. Conf. on Utility and Cloud Computing Companion 2018 (UCC Companion), 181–188. doi:https://doi.org/10.1109/UCC-Companion.2018.00054
Bajpai A (2021) Serverless Cold Starts - Mitigation Techniques. https://www.techtalksbyanvita.com/post/serverless-cold-starts-can-we-mitigate-these. Accessed 1 Mar 2023
Raza A, Matta I, Akhtar N, Kalavri V, Isahagian V (2021) SoK: Function-As-A-Service: From An Application Developer’s Perspective. J Syst Res. 1(1). https://doi.org/10.5070/SR31154815.
Baresi L and Filgueira Mendonça D (2019) Towards a Serverless Platform for Edge Computing. In: IEEE Int. Conf. on Fog Computing 2019 (ICFC’19), 1–10. https://doi.org/10.1109/ICFC.2019.00008
Szegedy C, Vanhoucke V, Ioffe S, Shlens J and Wojna Z (2016) Rethinking the Inception Architecture for Computer Vision. In: IEEE Conf. on Computer Vision and Pattern Recognition 2016, 2818–2826. https://doi.org/10.1109/CVPR.2016.308
Moreno-Vozmediano R. https://github.com/rmorvoz/FaaSim. Accessed 1 Mar 2023
Amazon Web Services, Inc. https://aws.amazon.com/lambda. Accessed 1 Mar 2023
Google. https://cloud.google.com/functions. Accessed 1 Mar 2023
Microsoft. https://azure.microsoft.com/services/functions/. Accessed 1 Mar 2023
Palade A, Kazmi A, Clarke S (2019) An Evaluation of Open Source Serverless Computing Frameworks Support at the Edge. In: IEEE World Congress on Services 2019, 206-211. https://doi.org/10.1109/SERVICES.2019.00057
Li J, Kulkarni S, Ramakrishnan K, Li D (2019) Understanding Open Source Serverless Platforms: Design Considerations and Performance. In: 5th Int. Workshop on Serverless Computing 2019 (WOSC ’19), 37–42. https://doi.org/10.1145/3366623.3368139
Mohanty S, Premsankar G, di Francesco M (2018) An Evaluation of Open Source Serverless Computing Frameworks. In: IEEE Int. Conf. on Cloud Computing Technology and Science 2018 (CloudCom), 115–120. doi:https://doi.org/10.1109/CloudCom2018.2018.00033
Ellis A. https://www.openfaas.com/. Accessed 1 Mar 2023
The Apache Software Foundation. https://openwhisk.apache.org/. Accessed 1 Mar 2023
Hendrickson S, Sturdevant S, Harter T, Venkataramani V, Arpaci-Dusseau A, Arpaci-Dusseau R (2016) Serverless computation with openLambda. In: 8th USENIX Conference on Hot Topics in Cloud Computing (HotCloud’16), USENIX Association, 33–39. https://dl.acm.org/doi/https://doi.org/10.5555/3027041.3027047
Kubeless (VMware Archive). https://github.com/vmware-archive/kubeless. Accessed 1 Mar 2023
The Kubernetes Authors. https://kubernetes.io/. Accessed 1 Mar 2023
The Knative Authors. https://knative.dev/. Accessed 1 Mar 2023
The Istio Authors. https://istio.io/. Accessed 1 Mar 2023
Díaz J. https://www.npmjs.com/package/serverless-plugin-warmup. Accessed 1 Mar 2023
Shahrad M et al (2020) Serverless in the wild: characterizing and optimizing the serverless workload at a large cloud provider. In: USENIX Annual Technical Conference 2020, 205–128. https://doi.org/10.5555/3489146.3489160
Fuerst A, Sharma P (2021) FaasCache: keeping serverless computing alive with greedy-dual caching. In: 26th ACM Int. Conf. on Architectural Support for Programming Languages and Operating Systems 2021(ASPLOS’21), 386–400. doi:https://doi.org/10.1145/3445814.3446757
Roy R, Patel T, Tiwari D (2022) IceBreaker: warming serverless functions better with heterogeneity. In: 27th ACM Int. Conf. on Architectural Support for Programming Languages and Operating Systems 2022 (ASPLOS 2022), 753–767. doi:https://doi.org/10.1145/3503222.3507750
Amazon Web Services, Inc, “Configuring provisioned concurrency,” https://docs.aws.amazon.com/lambda/latest/dg/provisioned-concurrency.html. Accessed 1 Mar 2023
Microsoft Azure, “Azure Functions Premium Plan”, https://learn.microsoft.com/en-us/azure/azure-functions/functions-premium-plan. Accessed 1 Mar 2023
The Apache Software Foundation, “OpenWhisk Actions”, https://apache.googlesource.com/openwhisk/+/HEAD/docs/actions.md. Accessed 1 Mar 2023
Silva P, Fireman D, Emmanuel Pereira T (2020) Prebaking Functions to Warm the Serverless Cold Start. In: 21st Int. Middleware Conference 2020, pp. 1–13. doi:https://doi.org/10.1145/3423211.3425682
Agarwal S, Rodriguez M, Buyya R (2021) A Reinforcement Learning Approach to Reduce Serverless Function Cold Start Frequency. I: 21st International Symposium on Cluster, Cloud and Internet Computing 2021 (CCGrid), 797–803. doi:https://doi.org/10.1109/CCGrid51090.2021.00097
Benedetti P, Femminella M, Reali G, Steenhaut K (2021) Experimental Analysis of the Application of Serverless Computing to IoT Platforms. Sensors, 21(3). https://doi.org/10.3390/s21030928
Aslanpour M et al. (2021) Serverless Edge Computing: Vision and Challenges. In: Australasian Computer Science Week Multiconference 2021 (ACSW '21), 1–10. https://doi.org/10.1145/3437378.3444367
Baresi L, Quattrocchi G (2021) PAPS: A Serverless Platform for Edge Computing Infrastructures. Frontiers in Sustainable Cities 3. https://doi.org/10.3389/frsc.2021.690660
Gadepalli P, Peach G, Cherkasova L, Aitken R, Parmer G (2019) Challenges and Opportunities for Efficient Serverless Computing at the Edge. In: 38th Symposium on Reliable Distributed Systems 2019 (SRDS), 261–266. https://doi.org/10.1109/SRDS47363.2019.00036
Xie Q, Tang S, Qiao H, Zhu F, Yu R, Huang T (2021) When Serverless Computing Meets Edge Computing: Architecture, Challenges, and Open Issues. IEEE Wireless Commun 28(5):126–133. https://doi.org/10.1109/MWC.001.2000466
Malishev N (2019) AWS Lambda Cold Start Language Comparisons, 2019 edition. https://levelup.gitconnected.com/aws-lambda-cold-start-language-comparisons-2019-edition-%EF%B8%8F-1946d32a0244. Accessed 1 Mar 2023
Roberts M (2020) Analyzing Cold Start latency of AWS Lambda. https://blog.symphonia.io/posts/2020-06-30_analyzing_cold_start_latency_of_aws_lambda. Accessed 1 Mar 2023
Moreno-Vozmediano R, Montero R, Llorente I (2012) IaaS Cloud Architecture: From Virtualized Datacenters to Federated Cloud Infrastructures. Computer 45(12):65–72. https://doi.org/10.1109/MC.2012.76
Moreno-Vozmediano R et al (2016) BEACON: A Cloud Network Federation Framework. In: Celesti A, Leitner P (eds) Advances in Service-Oriented and Cloud Computing. ESOCC 2015. Communications in Computer and Information Science, vol 567. Springer, Cham. https://doi.org/10.1007/978-3-319-33313-7_25
Silberman N, Guadarrama S (2016) TensorFlow-Slim image classification model library. https://github.com/tensorflow/models/tree/master/research/slim. Accessed 1 Mar 2023
Russakovsky O et al (2015) ImageNet Large Scale Visual Recognition Challenge. Int J Comput Vision 115:211–252. https://doi.org/10.1007/s11263-015-0816-y
WekaDeeplearning4j (2019) IMAGENET 1000 Class List. https://deeplearning.cms.waikato.ac.nz/user-guide/class-maps/IMAGENET/. Accessed 1 Mar 2023
Ivanovic B, Ivanovic Z (2017) How to Deploy Deep Learning Models with AWS Lambda and Tensorflow. https://aws.amazon.com/blogs/machine-learning/how-to-deploy-deep-learning-models-with-aws-lambda-and-tensorflow. Accessed 1 Mar 2023
Corrado A (2020) Kaggle Animals-10 dataset. https://www.kaggle.com/datasets/alessiocorrado99/animals10. Accessed 1 Mar 2023
Amazon Web Services, Inc, Lambda execution environment. https://docs.aws.amazon.com/lambda/latest/dg/lambda-runtime-environment.html. Accessed 1 Mar 2023