Cold Start Latency Optimization Strategies for Function as a Service Platforms

Tingjie Chen; Jianbo Ding

doi:10.54097/ya09a396

Authors

Tingjie Chen
Jianbo Ding

DOI:

https://doi.org/10.54097/ya09a396

Keywords:

Function as a Service, Serverless computing, Cold start latency, Pre-warming, Checkpoint-restore, Lightweight virtualization, WebAssembly, Scheduling optimization

Abstract

Function as a Service (FaaS) has established itself as the dominant delivery model within the serverless computing (SC) ecosystem, enabling developers to deploy stateless, event-driven workloads without provisioning or managing any underlying server infrastructure. Despite substantial operational advantages in cost granularity and scaling automation, FaaS platforms are subject to a persistent performance bottleneck known as cold start latency, which occurs whenever a new execution environment must be initialized from scratch before an incoming function invocation can be served. Cold start penalties range from tens of milliseconds for lightweight runtimes to several seconds for applications executing on the Java Virtual Machine (JVM), producing direct violations of service-level objectives (SLOs) in latency-sensitive production deployments. This paper reviews optimization strategies for cold start latency across four principal categories: pre-warming and keep-alive policies, snapshot-based checkpoint-restore techniques, lightweight virtualization and isolation mechanisms including WebAssembly (Wasm), and scheduling and resource management strategies. Machine learning (ML) is examined as a cross-cutting enabler for predictive and adaptive mitigation. This review synthesizes the current state of understanding, characterizes trade-offs among competing strategies, and identifies open challenges including snapshot staleness management, isolation-speed tensions, and edge deployment constraints.

Downloads

Download data is not yet available.

References

[1] Leitner, P., Wittern, E., Spillner, J., & Hummer, W. (2019). A mixed-method empirical study of function-as-a-service software development in industrial practice. Journal of Systems and Software, 149, 340-359.

[2] Eismann, S., Scheuner, J., Van Eyk, E., Schwinger, M., Grohmann, J., Herbst, N., ... & Iosup, A. (2021). The state of serverless applications: Collection, characterization, and community consensus. IEEE Transactions on Software Engineering, 48(10), 4152-4166.

[3] Cordingly, R., Yu, H., Hoang, V., Perez, D., Foster, D., Sadeghi, Z., ... & Lloyd, W. J. (2020, August). Implications of programming language selection for serverless data processing pipelines. In 2020 IEEE Intl Conf on Dependable, Autonomic and Secure Computing, Intl Conf on Pervasive Intelligence and Computing, Intl Conf on Cloud and Big Data Computing, Intl Conf on Cyber Science and Technology Congress (DASC/PiCom/CBDCom/CyberSciTech) (pp. 704-711). IEEE.

[4] Jonas, E., Schleier-Smith, J., Sreekanti, V., Tsai, C. C., Khandelwal, A., Pu, Q., ... & Patterson, D. A. (2019). Cloud programming simplified: A berkeley view on serverless computing. arXiv preprint arXiv:1902.03383.

[5] Mohan, A., Sane, H., Doshi, K., Edupuganti, S., Nayak, N., & Sukhomlinov, V. (2019). Agile cold starts for scalable serverless. In 11th USENIX Workshop on Hot Topics in Cloud Computing (HotCloud 19).

[6] Varshney, R. P., & Sharma, D. K. (2020, October). Cold start in function as a service: a systematic study, analysis and evaluation. In International Conference on Futuristic Trends in Networks and Computing Technologies (pp. 337-349). Singapore: Springer Singapore.

[7] Shahrad, M., Balkind, J., & Wentzlaff, D. (2019, October). Architectural implications of function-as-a-service computing. In Proceedings of the 52nd annual IEEE/ACM international symposium on microarchitecture (pp. 1063-1075).

[8] Wen, J., Chen, Z., Jin, X., & Liu, X. (2023). Rise of the planet of serverless computing: A systematic review. ACM Transactions on Software Engineering and Methodology, 32(5), 1-61.

[9] Lin, P. M., & Glikson, A. (2019). Mitigating cold starts in serverless platforms: A pool-based approach. arXiv preprint arXiv:1903.12221.

[10] Tariq, A., Pahl, A., Nimmagadda, S., Rozner, E., & Lanka, S. (2020, October). Sequoia: Enabling quality-of-service in serverless computing. In Proceedings of the 11th ACM symposium on cloud computing (pp. 311-327).

[11] Mendki, P. (2020, October). Evaluating webassembly enabled serverless approach for edge computing. In 2020 IEEE Cloud Summit (pp. 161-166). IEEE.

[12] Ding, G., Yang, S., Lin, H., Chen, Z., & Yang, J. S. (2026). LLM-Driven Adaptive Cloud Resource Scheduling: Bridging Reasoning Intelligence with Optimization Guarantees. IEEE Open Journal of the Computer Society.

[13] Eismann, S., Scheuner, J., Van Eyk, E., Schwinger, M., Grohmann, J., Herbst, N., ... & Iosup, A. (2020). A review of serverless use cases and their characteristics. arXiv preprint arXiv:2008.11110.

[14] Li, Y., Lin, Y., Wang, Y., Ye, K., & Xu, C. (2022). Serverless computing: state-of-the-art, challenges and opportunities. IEEE Transactions on Services Computing, 16(2), 1522-1539.

[15] Shahrad, M., Fonseca, R., Goiri, I., Chaudhry, G., Batum, P., Cooke, J., ... & Bianchini, R. (2020). Serverless in the wild: Characterizing and optimizing the serverless workload at a large cloud provider. In 2020 USENIX annual technical conference (USENIX ATC 20) (pp. 205-218).

[16] Kulkarni, V., Reddy, N., Khare, T., Mohan, H., Murali, J., Balajee, S., ... & Simmhan, Y. (2024, May). XFBench: a cross-cloud benchmark suite for evaluating faas workflow platforms. In 2024 IEEE 24th International Symposium on Cluster, Cloud and Internet Computing (CCGrid) (pp. 543-556). IEEE.

[17] Agache, A., Brooker, M., Iordache, A., Liguori, A., Neugebauer, R., Piwonka, P., & Popa, D. M. (2020). Firecracker: Lightweight virtualization for serverless applications. In 17th USENIX symposium on networked systems design and implementation (NSDI 20) (pp. 419-434).

[18] Zhao, T., Hall, M., Johansen, H., & Williams, S. (2021, February). Improving communication by optimizing on-node data movement with data layout. In Proceedings of the 26th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming (pp. 304-317).

[19] Fuerst, A., & Sharma, P. (2021, April). Faascache: keeping serverless computing alive with greedy-dual caching. In Proceedings of the 26th ACM international conference on architectural support for programming languages and operating systems (pp. 386-400).

[20] Shillaker, S., & Pietzuch, P. (2020). Faasm: Lightweight isolation for efficient stateful serverless computing. In 2020 USENIX Annual Technical Conference (USENIX ATC 20) (pp. 419-433).

[21] Sreekanti, V., Wu, C., Lin, X. C., Schleier-Smith, J., Faleiro, J. M., Gonzalez, J. E., ... & Tumanov, A. (2020). Cloudburst: Stateful functions-as-a-service. arXiv preprint arXiv:2001.04592.

[22] Bermbach, D., Karakaya, A. S., & Buchholz, S. (2020, March). Using application knowledge to reduce cold starts in FaaS services. In Proceedings of the 35th annual ACM symposium on applied computing (pp. 134-143).

[23] Sui, Y., Yu, H., Hu, Y., Li, J., & Wang, H. (2024, November). Pre-warming is not enough: Accelerating serverless inference with opportunistic pre-loading. In Proceedings of the 2024 ACM Symposium on Cloud Computing (pp. 178-195).

[24] Aslanpour, M. S., Toosi, A. N., Cicconetti, C., Javadi, B., Sbarski, P., Taibi, D., ... & Dustdar, S. (2021, February). Serverless edge computing: vision and challenges. In Proceedings of the 2021 Australasian computer science week multiconference (pp. 1-10).

[25] Pfandzelter, T., & Bermbach, D. (2020, April). tinyfaas: A lightweight faas platform for edge environments. In 2020 IEEE International Conference on Fog Computing (ICFC) (pp. 17-24). IEEE.

[26] Golec, M., Walia, G. K., Kumar, M., Cuadrado, F., Gill, S. S., & Uhlig, S. (2024). Cold start latency in serverless computing: A systematic review, taxonomy, and future directions. ACM Computing Surveys, 57(3), 1-36.

[27] Daw, N., Bellur, U., & Kulkarni, P. (2021, November). Speedo: Fast dispatch and orchestration of serverless workflows. In Proceedings of the ACM Symposium on Cloud Computing (pp. 585-599).

[28] Anjali, Caraza-Harter, T., & Swift, M. M. (2020, March). Blending containers and virtual machines: a study of firecracker and gVisor. In Proceedings of the 16th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments (pp. 101-113).

[29] Das, A., Imai, S., Patterson, S., & Wittie, M. P. (2020, May). Performance optimization for edge-cloud serverless platforms via dynamic task placement. In 2020 20th IEEE/ACM International Symposium on Cluster, Cloud and Internet Computing (CCGRID) (pp. 41-50). IEEE.

[30] Li, H., Yuan, Y., Du, R., Ma, K., Liu, L., & Hsu, W. (2020). {DADI}:{Block-Level} Image Service for Agile and Elastic Application Deployment. In 2020 USENIX Annual Technical Conference (USENIX ATC 20) (pp. 727-740).

[31] Vahidinia, P., Farahani, B., & Aliee, F. S. (2020, August). Cold start in serverless computing: Current trends and mitigation strategies. In 2020 International Conference on Omni-layer Intelligent Systems (COINS) (pp. 1-7). IEEE.

[32] Ménétrey, J., Pasin, M., Felber, P., & Schiavoni, V. (2022, July). Webassembly as a common layer for the cloud-edge continuum. In Proceedings of the 2nd Workshop on Flexible Resource and Application Management on the Edge (pp. 3-8).

[33] Barrak, A., Petrillo, F., & Jaafar, F. (2022). Serverless on machine learning: A systematic mapping study. IEEE Access, 10, 99337-99352.

[34] Toutain, L., Elloumi, O., Liang, S., van der Schaaf, H., De Lathouwer, B., Minaburo, A., & Raggett, D. (2023). Application Layer Protocols. In Springer Handbook of Internet of Things (pp. 481-507). Cham: Springer International Publishing.

[35] Boza, E. F., Andrade, X., Cedeno, J., Murillo, J., Aragon, H., Abad, C. L., & Abad, A. G. (2020). On implementing autonomic systems with a serverless computing approach: The case of self-partitioning cloud caches. Computers, 9(1), 14.

[36] Chen, Z., Wang, Y., & Zhao, X. (2025). Responsible Generative AI: Governance Challenges and Solutions in Enterprise Data Clouds. Journal of Computing and Electronic Information Management, 18(3), 59-65.

[37] Rausch, T., Hummer, W., Muthusamy, V., Rashed, A., & Dustdar, S. (2019). Towards a serverless platform for edge {AI}. In 2nd USENIX Workshop on Hot Topics in Edge Computing (HotEdge 19).

[38] Mampage, A., Karunasekera, S., & Buyya, R. (2022). A holistic view on resource management in serverless computing environments: Taxonomy and future directions. ACM Computing Surveys (CSUR), 54(11s), 1-36.

[39] Eismann, S., Grohmann, J., Van Eyk, E., Herbst, N., & Kounev, S. (2020, April). Predicting the costs of serverless workflows. In Proceedings of the ACM/SPEC international conference on performance engineering (pp. 265-276).

[40] Besozzi, V., Della Bartola, M., Dazzi, P., & Danelutto, M. (2025). High-Performance Serverless Computing: A Systematic Literature Review on Serverless for HPC, AI, and Big Data. IEEE Access, 13, 195611-195656.

[41] Mahmoudi, N., & Khazaei, H. (2020). Performance modeling of serverless computing platforms. IEEE Transactions on Cloud Computing, 10(4), 2834-2847.

[42] Vahidinia, P., Farahani, B., & Aliee, F. S. (2022). Mitigating cold start problem in serverless computing: A reinforcement learning approach. IEEE Internet of Things Journal, 10(5), 3917-3927.

[43] Gunasekaran, J. R., Thinakaran, P., Nachiappan, N. C., Kandemir, M. T., & Das, C. R. (2020, December). Fifer: Tackling resource underutilization in the serverless era. In Proceedings of the 21st International Middleware Conference (pp. 280-295).

[44] Ghorbian, M., & Ghobaei-Arani, M. (2024). A survey on the cold start latency approaches in serverless computing: an optimization-based perspective. Computing, 106(11), 3755-3809.

[45] Schleier-Smith, J. M. (2022). Understanding and Exploring Serverless Cloud Computing. University of California, Berkeley.

[46] Gias, A. U., Casale, G., & Woodside, M. (2019, July). ATOM: Model-driven autoscaling for microservices. In 2019 IEEE 39th International Conference on Distributed Computing Systems (ICDCS) (pp. 1994-2004). IEEE.

[47] Wang, R., Yu, G., Casale, G., Chen, P., & Filieri, A. (2026). Fine-grained Tracing for Performance Anomaly Diagnosis of Serverless Functions. ACM Transactions on Autonomous and Adaptive Systems.

[48] Cadden, J., Unger, T., Awad, Y., Dong, H., Krieger, O., & Appavoo, J. (2020, April). SEUSS: skip redundant paths to make serverless fast. In Proceedings of the Fifteenth European Conference on Computer Systems (pp. 1-15).

[49] Chen, Y., Liu, B., Lin, W., Guo, Y., & Peng, Z. (2025). CASR: Optimizing cold start and resources utilization in serverless computing. Future Generation Computer Systems, 170, 107851.

[50] Holmes, B., Dinis, B., Honcharuk, L., Fried, J., & Belay, A. (2025). Taming Serverless Cold Starts Through OS Co-Design. arXiv preprint arXiv:2509.14292.

[51] Wang, S. (2021). Thin serverless functions with graalvm native image.

[52] Agarwal, S., Rodriguez, M. A., & Buyya, R. (2025). Learning in Serverless Computing. Computational Intelligence and Data Analytics: Proceedings of ICCIDA 2024, 1.

[53] Kim, S. J., You, M., Kim, B. J., & Shin, S. (2023, October). Cryonics: Trustworthy Function-as-a-Service using Snapshot-based Enclaves. In Proceedings of the 2023 ACM Symposium on Cloud Computing (pp. 528-543).

[54] Ustiugov, D., Petrov, P., Kogias, M., Bugnion, E., & Grot, B. (2021, April). Benchmarking, analysis, and optimization of serverless function snapshots. In Proceedings of the 26th ACM international conference on architectural support for programming languages and operating systems (pp. 559-572).

[55] Jain, S. M. (2020). Linux Containers and Virtualization. A Kernel Perspective, 2020-10.

[56] Gackstatter, P., Frangoudis, P. A., & Dustdar, S. (2022, May). Pushing serverless to the edge with webassembly runtimes. In 2022 22nd IEEE International Symposium on Cluster, Cloud and Internet Computing (CCGrid) (pp. 140-149). IEEE.

[57] Ramachandran, U., Gupta, H., Hall, A., Saurez, E., & Xu, Z. (2019, July). Elevating the edge to be a peer of the cloud. In 2019 IEEE 12th International Conference on Cloud Computing (CLOUD) (pp. 17-24). IEEE.

[58] Panda, A., & Sarangi, S. R. (2024). Faasctrl: A comprehensive-latency controller for serverless platforms. IEEE Transactions on Cloud Computing, 12(4), 1328-1343.

[59] Shi, J., Gu, J., Xia, Y., & Chen, H. (2025). Serverless functions made confidential and efficient with split containers. In 34th USENIX Security Symposium (USENIX Security 25) (pp. 1091-1110).

[60] Verma, P., Goel, P., & Rani, N. (2024, April). A review: Cold start latency in serverless computing. In 2024 Sixth International Conference on Computational Intelligence and Communication Technologies (CCICT) (pp. 141-148). IEEE.

[61] Liu, X., Wen, J., Chen, Z., Li, D., Chen, J., Liu, Y., ... & Jin, X. (2023). Faaslight: General application-level cold-start latency optimization for function-as-a-service in serverless computing. ACM Transactions on Software Engineering and Methodology, 32(5), 1-29.

[62] Zhang, Y., & Jacobsen, H. A. (2025, December). Mocha: Scalable and Compliant Function Scheduling for Federated Serverless Computing. In Proceedings of the 26th International Middleware Conference (pp. 398-412).

[63] Kaffes, K., Yadwadkar, N. J., & Kozyrakis, C. (2019, November). Centralized core-granular scheduling for serverless functions. In Proceedings of the ACM symposium on cloud computing (pp. 158-164).

[64] Brown, B. (2025). Serverless Architectures for Scalable Cloud-Based Microservices: Performance and Cost Trade-Offs.

[65] Baresi, L., Hu, D. Y. X., Quattrocchi, G., & Terracciano, L. (2022, May). NEPTUNE: Network-and GPU-aware management of serverless functions at the edge. In Proceedings of the 17th Symposium on Software Engineering for Adaptive and Self-Managing Systems (pp. 144-155).