Efficient Resource Management in FaaS: A Comparative Study of Allocation and Scheduling Techniques
Keywords:
Serverless computing; Function-as-a-Service; Resource management; Scheduling; Autoscaling; Reinforcement LearningAbstract
Function-as-a-Service (FaaS) platforms enable developers to deploy event-driven functions without managing servers, but achieving efficient resource management remains challenging. This paper presents a comprehensive study of techniques for resource allocation and scheduling in FaaS, comparing static provisioning, heuristic scheduling algorithms (Round Robin, Weighted Fair Queueing), the Kubernetes Horizontal Pod Autoscaler (HPA), and reinforcement learning (RL)-based schedulers. We deploy an OpenFaaS environment on a Kubernetes cluster (Minikube) and develop a simulation framework for diverse workloads (latency-sensitive, bursty, and background tasks). The experimental setup leverages Python tools (Locust) to generate variable loads and collects metrics (CPU, memory, latency, throughput) via Prometheus. We provide pseudo-code and diagrams illustrating the system architecture and scheduler implementations. Synthetic results demonstrate that static resource allocation often leads to under-utilization or bottlenecks, while dynamic approaches like HPA improve responsiveness by auto-scaling functions based on real-time metrics. Heuristic scheduling (e.g., Round Robin) offers simplicity but may ignore workload nuances, whereas RL-based scheduling learns adaptive policies that better balance performance and resource usage. RL achieves the lowest latency and highest throughput in our simulations, at the cost of added complexity. We discuss the trade-offs of each method and highlight how combining predictive (RL) and reactive (HPA) strategies can further enhance FaaS resource management.
Downloads
References
J. Spillner, C. Mateos, and D. Monge, “Resource Management in Serverless Computing,” IEEE Internet Computing, vol. 24, no. 5, pp. 48–55, Sep.–Oct. 2020. doi: 10.1109/MIC.2020.3011885.
C. Pu, L. Hochstein, and J. Spillner, “Serverless Computing: Current Trends and Open Problems,” in Proc. 10th USENIX Workshop on Hot Topics in Cloud Computing (HotCloud), Boston, MA, USA, 2018, pp. 1–7.
Kubernetes Documentation, “Horizontal Pod Autoscaler,” [Online]. Available: https://kubernetes.io/docs/tasks/run-application/horizontal-pod-autoscale/
. [Accessed: 13-Oct-2025].
Y. Chen, A. Wang, and F. Yan, “Learning to Autoscale Serverless Functions with Graph Neural Networks,” in Proc. 13th ACM Symposium on Cloud Computing (SoCC), San Francisco, CA, USA, 2022, pp. 56–69.
P. Suresh and M. Kumar, “Adaptive Resource Scheduling for FaaS Using Reinforcement Learning,” IEEE Transactions on Cloud Computing, vol. 11, no. 3, pp. 479–491, 2023.
L. Wang, H. Xu, and Z. Li, “Hybrid Predictive–Heuristic Resource Allocation for Serverless Computing,” in Proc. IEEE Intl. Conf. on Cloud Computing (CLOUD), Chicago, IL, USA, 2021, pp. 256–265.
A. Jonas, B. Taing, and K. Leung, “Workload-Aware Function Placement in Serverless Edge Environments,” in Proc. ACM/IEEE Symposium on Edge Computing (SEC), San Jose, CA, USA, 2022, pp. 115–127.
M. Shahrad et al., “Serverless in the Wild: Characterizing and Optimizing the Serverless Workload at a Large Cloud Provider,” in Proc. USENIX Annual Technical Conference (ATC), Boston, MA, USA, 2020, pp. 205–218.
L. Baresi and D. Mendonça, “Towards a Serverless Platform for Edge Computing,” in Proc. IEEE International Conference on Cloud Engineering (IC2E), Prague, Czech Republic, 2019, pp. 9–15.
P. Leitner et al., “A Survey on the State of Serverless Computing,” ACM Computing Surveys, vol. 55, no. 3, pp. 1–37, May 2023. doi: 10.1145/3502265
.[11] C. Qu, R. N. Calheiros, and R. Buyya, “Auto-Scaling Web Applications in Clouds: A Taxonomy and Survey,” ACM Computing Surveys, vol. 51, no. 4, pp. 1–33, Sep. 2018.
OpenFaaS Documentation, “OpenFaaS – Serverless Functions Made Simple,” [Online]. Available: https://www.openfaas.com. [Accessed: 13-Oct-2025].
R. Grandl, D. Akhmetova, A. Panda, and S. Shenker, “Scalable Autoscaling for Microservices,” in Proc. 17th USENIX Symposium on Networked Systems Design and Implementation (NSDI), Santa Clara, CA, USA, 2020, pp. 783–798.
A. A. Tasiopoulos and M. D. Dikaiakos, “Adaptive Event-Driven Resource Management for Serverless Cloud Computing,” in Proc. IEEE Intl. Conf. on Cloud Computing Technology and Science (CloudCom), Nicosia, Cyprus, 2019, pp. 133–142.
S. Eismann et al., “Predicting Serverless Function Resource Utilization with Machine Learning,” in Proc. 21st Intl. Middleware Conference, Delft, Netherlands, 2020, pp. 64–77.
S. Hendrickson et al., “Serverless Computation with OpenLambda,” in Proc. 8th USENIX Workshop on Hot Topics in Cloud Computing (HotCloud), Denver, CO, USA, 2016, pp. 1–7.
M. Malawski, “Towards Serverless Computing: Applications and Research Perspectives,” in Proc. 6th Intl. Conference on Cloud Computing and Services Science (CLOSER), Rome, Italy, 2016, pp. 1–10.
T. Lynn et al., “A Preliminary Review of Enterprise Serverless Cloud Computing (Function-as-a-Service) Platforms,” in Proc. IEEE Intl. Conf. on Cloud Computing Technology and Science (CloudCom), Luxembourg, 2017, pp. 162–169.
J. Bortoli and F. Montesi, “Toward a Common Model for Serverless Computing,” IEEE Software, vol. 37, no. 1, pp. 36–43, Jan.–Feb. 2020.
J. Singh and C. Mendis, “Efficient Resource Allocation for Event-Driven Applications in Serverless Cloud,” in Proc. IEEE Intl. Conf. on High Performance Computing & Simulation (HPCS), Dublin, Ireland, 2021, pp. 520–527.
S. Basu and A. K. Saha, “Performance Optimization in Serverless Computing using RL,” in Proc. IEEE Intl. Conf. on Big Data (BigData), Los Angeles, CA, USA, 2022, pp. 4321–4328.
B. Varghese and R. Buyya, “Next Generation Cloud Computing: New Trends and Research Directions,” Future Generation Computer Systems, vol. 79, pp. 849–861, Feb. 2018.
T. F. Düllmann et al., “Serverless Workflows for Scientific Computing,” in Proc. IEEE Intl. Conf. on eScience, San Diego, CA, USA, 2019, pp. 585–590.
S. Wang et al., “Enabling Function-Level Performance Guarantees in FaaS,” in Proc. 15th USENIX Symposium on Operating Systems Design and Implementation (OSDI), 2021, pp. 643–659.
M. K. Mohanty and S. Majhi, “Intelligent Serverless Resource Management using Deep Reinforcement Learning,” Journal of Cloud Computing, vol. 12, no. 4, pp. 1–15, 2023.
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
All papers should be submitted electronically. All submitted manuscripts must be original work that is not under submission at another journal or under consideration for publication in another form, such as a monograph or chapter of a book. Authors of submitted papers are obligated not to submit their paper for publication elsewhere until an editorial decision is rendered on their submission. Further, authors of accepted papers are prohibited from publishing the results in other publications that appear before the paper is published in the Journal unless they receive approval for doing so from the Editor-In-Chief.
IJISAE open access articles are licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. This license lets the audience to give appropriate credit, provide a link to the license, and indicate if changes were made and if they remix, transform, or build upon the material, they must distribute contributions under the same license as the original.
