Dmitrii Ustiugov
ABSTRACT
Serverless is an increasingly popular cloud computing paradigm that has stimulated new systems research opportunities. However, developing and evaluating serverless systems in a research setting (i.e., "in-vitro", without access to a large-scale production cluster and real workloads) is challenging yet vital for innovation. Recently, several serverless providers have released production traces consisting of large sets of functions with their invocation inter-arrival time, execution time, and memory footprint distributions. However, executing the workload synthesized from these traces requires a massive cluster, making experiments expensive and time-consuming.
In this work, we show how to use the data available in production traces to construct workload summaries of configurable scales that remain highly representative of the original trace characteristics and can be used to evaluate serverless systems in-vitro. Compared to random sampling of functions from the original trace, our method can generate summaries of up to 10X higher representativity, measured as the average of the Wasserstein distances of the distributions of interest (e.g., function execution time and invocation inter-arrival time) from the respective distributions in the original trace. We release our toolchain that enables researchers to synthesize representative workload summaries and show how it can be used to evaluate the performance of serverless systems at diverse load scale factors.
- Additional Autoscaling Configuration for Knative Pod Autoscaler. Available at https://knative.dev/docs/serving/autoscaling/kpa-specific/#stable-window.Google Scholar
- Fission: Open Source, Kubernetes-Native Serverless Framework. Available at https://fission.io.Google Scholar
- Fn project. Available at https://fnproject.io.Google Scholar
- Knative. Available at https://knative.dev.Google Scholar
- Prometheus. Available at https://prometheus.io.Google Scholar
- Transaction Processing Performance Council. Available at https://www.tpc.org.Google Scholar
- What are Industry 4.0, the Fourth Industrial Revolution, and 4IR? Available at https://www.mckinsey.com/featured-insights/mckinsey-explainers/what-are-industry-4-0-the-fourth-industrial-revolution-and-4ir.Google Scholar
- Apache. OpenWhisk. Available at https://openwhisk.apache.org/.Google Scholar
- Datadog. The State of Serverless 2021. Available at https://www.datadoghq.com/state-of-serverless-2021.Google Scholar
- Deep, S., Gruenheid, A., Koutris, P., Naughton, J. F., and Viglas, S. Comprehensive and Efficient Workload Compression. Proc. VLDB Endow. 14, 3 (2020), 418--430.Google Scholar
- Duplyakin, D., Ricci, R., Maricq, A., Wong, G., Duerig, J., Eide, E., Stoller, L., Hibler, M., Johnson, D., Webb, K., Akella, A., Wang, K.-C., Ricart, G., Landweber, L., Elliott, C., Zink, M., Cecchet, E., Kar, S., and Mishra, P. The Design and Operation of CloudLab. In Proceedings of the 2019 USENIX Annual Technical Conference (ATC) (2019), pp. 1--14.Google Scholar
- Fuerst, A., and Sharma, P. FaasCache: Keeping Serverless Computing Alive with Greedy-Dual Caching. In Proceedings of the 26th International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS-XXVI) (2021), pp. 386--400.Google ScholarDigital Library
- Hendrickson, S., Sturdevant, S., Harter, T., Venkataramani, V., Arpaci-Dusseau, A. C., and Arpaci-Dusseau, R. H. Serverless Computation with OpenLambda. In 8th USENIX Workshop on Hot Topics in Cloud Computing (HotCloud) (2016).Google Scholar
- Kaffes, K., Yadwadkar, N. J., and Kozyrakis, C. Hermod: Principled and Practical Scheduling for Serverless Functions. In Proceedings of the 2022 ACM Symposium on Cloud Computing (SOCC) (2022), pp. 289--305.Google ScholarDigital Library
- Kogias, M., Mallon, S., and Bugnion, E. Lancet: A Self-Correcting Latency Measuring Tool. In Proceedings of the 2019 USENIX Annual Technical Conference (ATC) (2019), pp. 881--896.Google Scholar
- Kubeless. Kubeless: The Kubernetes Native Serverless Framework. Available at https://kubeless.io.Google Scholar
- Leverich, J., and Kozyrakis, C. Reconciling High Server Utilization and Sub-Millisecond Quality-of-service. In Proceedings of the 2014 EuroSys Conference (2014), pp. 4:1--4:14.Google Scholar
- Li, Z., Zheng, L., Zhong, Y., Liu, V., Sheng, Y., Jin, X., Huang, Y., Chen, Z., Zhang, H., Gonzalez, J. E., and Stoica, I. AlpaServe: Statistical Multiplexing with Model Parallelism for Deep Learning Serving. In 17th USENIX Symposium on Operating Systems Design and Implementation (OSDI) (2023), USENIX Association.Google Scholar
- Liang, M., Fu, W., Feng, L., Lin, Z., Panakanti, P., Zheng, S., Sridharan, S., and Delimitrou, C. Mystique: Enabling Accurate and Scalable Generation of Production AI Benchmarks. In Proceedings of the 50th Annual International Symposium on Computer Architecture, ISCA 2023, Orlando, FL, USA, June 17--21, 2023 (2023), Y. Solihin and M. A. Heinrich, Eds., ACM, pp. 37:1--37:13.Google ScholarDigital Library
- Panda, R., Zheng, X., Gerstlauer, A., and John, L. K. CAMP: Accurate Modeling of Core and Memory Locality for Proxy Generation of Big-Data Applications. In 2018 Design, Automation & Test in Europe Conference & Exhibition, DATE 2018, Dresden, Germany, March 19--23, 2018 (2018), J. Madsen and A. K. Coskun, Eds., IEEE, pp. 337--342.Google ScholarCross Ref
- Saxena, D., Ji, T., Singhvi, A., Khalid, J., and Akella, A. Memory Deduplication for Serverless Computing with Medes. In Proceedings of the 2022 EuroSys Conference (2022), pp. 714--729.Google Scholar
- Shahrad, M., Fonseca, R., Goiri, I., Chaudhry, G., Batum, P., Cooke, J., Laureano, E., Tresness, C., Russinovich, M., and Bianchini, R. Serverless in the Wild: Characterizing and Optimizing the Serverless Workload at a Large Cloud Provider. In Proceedings of the 2020 USENIX Annual Technical Conference (ATC) (2020), pp. 205--218.Google Scholar
- Singhvi, A., Balasubramanian, A., Houck, K., Shaikh, M. D., Venkataraman, S., and Akella, A. Atoll: A Scalable Low-Latency Serverless Platform. In Proceedings of the 2021 ACM Symposium on Cloud Computing (SOCC) (2021), p. 138--152.Google ScholarDigital Library
- Ustiugov, D., Amariucai, T., and Grot, B. Analyzing Tail Latency in Serverless Clouds with STeLLAR. In Proceedings of the 2021 IEEE International Symposium on Workload Characterization (IISWC) (2021), pp. 51--62.Google Scholar
- Ustiugov, D., Amariucai, T., and Grot, B. Analyzing Tail Latency in Serverless Clouds with STeLLAR. In IEEE International Symposium on Workload Characterization (IISWC) (2021).Google Scholar
- Ustiugov, D., Petrov, P., Kogias, M., Bugnion, E., and Grot, B. Benchmarking, Analysis, and Optimization of Serverless Function Snapshots. In Proceedings of the 26th International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS-XXVI) (2021), pp. 559--572.Google ScholarDigital Library
- Wang, A., Chang, S., Tian, H., Wang, H., Yang, H., Li, H., Du, R., and Cheng, Y. FaaSNet: Scalable and Fast Provisioning of Custom Serverless Container Runtimes at Alibaba Cloud Function Compute. In Proceedings of the 2021 USENIX Annual Technical Conference (ATC) (2021), pp. 443--457.Google Scholar
- Yu, Z., Eeckhout, L., Goswami, N., Li, T., John, L. K., Jin, H., Xu, C., and Wu, J. GPGPU-MiniBench: Accelerating GPGPU Micro-Architecture Simulation. IEEE Trans. Computers 64, 11 (2015), 3153--3166.Google Scholar
- Zhang, J., Jin, C., Huang, Y., Yi, L., Ding, Y., and Guo, F. KOLE: Breaking the Scalability Barrier for Managing Far Edge Nodes in Cloud. In Proceedings of the 2022 ACM Symposium on Cloud Computing (SOCC) (2022).Google ScholarDigital Library
Recommendations
Supporting Multi-Provider Serverless Computing on the Edge
ICPP Workshops '18: Workshop Proceedings of the 47th International Conference on Parallel ProcessingServerless computing has recently emerged as a new execution model for cloud computing, in which service providers offer compute runtimes, also known as Function-as-a-Service (FaaS) platforms, allowing users to develop, execute and manage application ...
Serverless Computing: Introduction and Research Challenges
Economics of Grids, Clouds, Systems, and ServicesAbstractServerless computing is a technology that offers the ability to create modular, highly-scalable, fault-tolerant applications, leveraging container-based virtualisation to deploy applications and services. It is revolutionising the way we think ...
Comments