END-TO-END PERFORMANCE MONITORING OF SERVERLESS ARCHITECTURES IN LATENCY-CRITICAL APPLICATION WORKLOADS

Espido Cristina Morales  Journal

END-TO-END PERFORMANCE MONITORING OF SERVERLESS ARCHITECTURES IN LATENCY-CRITICAL APPLICATION WORKLOADS

Authors

Espido Cristina Morales Journal

Application Reliability Engineer, Germany.

Keywords:

Serverless Computing, Performance Monitoring, End-to-End Monitoring, Latency-Critical Applications, Application Performance, Cloud Computing, Function As A Service (FaaS), Observability, Real-Time Monitoring, Serverless Architecture, Latency Optimization, Workload Profiling, Distributed Tracing, Performance Bottlenecks, Monitoring Tools, Resource Utilization, Service-Level Objectives (SLOs), Cold Start Latency, Telemetry Data, Monitoring Framework

Synopsis

Serverless computing has emerged as a disruptive cloud paradigm, enabling developers to deploy code without managing infrastructure. However, its benefits come with performance trade-offs, particularly in latency-critical application workloads. This paper explores the challenges and solutions associated with end-to-end performance monitoring of serverless systems, where traditional observability techniques may fail due to dynamic execution models, cold starts, and third-party dependencies. Through a synthesis of academic literature and empirical data from AWS Lambda, Azure Functions, and Google Cloud Functions, we present a comprehensive evaluation of latency bottlenecks. A novel hybrid monitoring framework is proposed that leverages OpenTelemetry tracing, machine learning-based anomaly detection, and orchestration-aware instrumentation. Our results demonstrate that strategic monitoring interventions can reduce latency violations by over 30% in multi-step workflows.

References

[1] Jia, Z., Witchel, E. (2021). Nightcore: efficient and scalable serverless computing for latency-sensitive, interactive microservices. In Proc. ACM Int. Conf. on Architectural Support for Programming Languages and Operating Systems, 194–208.

[2] Qiu, H., Jha, S., Banerjee, S.S., Patke, A., Wang, C. (2021). Is function-as-a-service a good fit for latency-critical services? Proc. ACM on Serverless Computing, 4(1), 1–13.

[3] Scheuner, J. (2022). Performance evaluation of serverless applications and infrastructures. PhD Thesis, Chalmers University.

[4] Yang, Y., Zhao, L., Li, Y., Zhang, H., Li, J., Zhao, M. (2022). INFless: a native serverless system for low-latency, high-throughput inference. Proc. ASPLOS, 492–508.

[5] Devalla, S. (2023). Adaptive predictive monitoring in enterprise serverless deployments. Int. J. of Computer Applications, 183(20), 34–41.

[6] Lyu, X., Cherkasova, L., Aitken, R., Parmer, G. (2022). Towards efficient processing of latency-sensitive serverless DAGs at the edge. In Proc. ACM/IFIP/USENIX Middleware, 118–130.

[7] Gummad, V. P. K. (2025). Flex gateway, service mesh, and advanced API management evolution. International Journal of Applied Mathematics, 38(9s), 2199–2206. https://doi.org/10.12732/ijam.v38i9s.1643

[8] Shahidi, N., Gunasekaran, J.R. (2021). Cross-platform performance evaluation of stateful serverless workflows. IEEE WoWMoM, 203–212.

[9] Ustiugov, D. (2022). Data-centric serverless cloud architecture. MSc Thesis, University of Edinburgh.

[10] Batchu, K.C. (2022). Serverless ETL with auto-scaling triggers: a performance-driven design. Int. J. of Computer Tech. and Engineering, 18(4), 87–96.

[11] Kulkarni, S.G., Liu, G., Ramakrishnan, K.K. (2019). Living on the edge: Serverless computing and the cost of failure resiliency. In IEEE SRDS, 95–104.

[12] Gunasekaran, J.R., Thinakaran, P. (2020). Fifer: Tackling resource underutilization in the serverless era. In ACM Middleware, 292–306.

[13] Pelle, I., Czentye, J., Dóka, J. (2019). Towards latency-sensitive cloud native applications. In IEEE CloudCom, 112–119.

[14] Kim, Y.K., HoseinyFarahabady, M.R. (2020). Automated fine-grained CPU cap control in serverless platforms. In IEEE IPDPSW,

143–150.

[15] Gadepalli, P.K., McBride, S., Peach, G. (2020). Sledge: A serverless-first, lightweight Wasm runtime for the edge. In Proc. ACM Middleware, 70–83.

[16] Mampage, A., Karunasekera, S., Buyya, R. (2022). A holistic view on resource management in serverless environments. ACM Comput. Surv., 55(1), 1–38.