Articles
| Open Access |
DOI: Startup Latency Analysis in Java Frameworks for Serverless AWS Lambda Deployments.
Maksimov Viacheslav Yurievich , Senior Software Engineer at AUTO1 IT Services SE & Co. KG, Germany, BerlinAbstract
Cold start latency in serverless computing, particularly in Java-based AWS Lambda functions, presents a significant challenge for latency-sensitive applications. This study investigates the performance characteristics of three modern Java frameworks - Spring Boot, Micronaut, and Quarkus - deployed on AWS Lambda using the ARM64 (Graviton2) architecture. It evaluates cold start latency across three deployment configurations: managed runtime (with and without SnapStart) and GraalVM native images. Metrics were collected at varying memory allocations using Java 21. Results show that Quarkus consistently outperforms others in cold start latency on standard JVM, while SnapStart and GraalVM significantly reduce the number of cold starts and achieve sub-second latency, respectively. We discuss the implications of these findings for choosing a Java framework and runtime strategy on AWS Lambda, considering the trade-offs in deployment time, complexity, and performance. The paper concludes with recommendations for leveraging SnapStart and native images to mitigate cold start issues in Java serverless applications on ARM64.
Keywords
AWS Lambda, Cold start, SnapStart, GraalVM, ARM64, Java
References
Dittakavi, R.S.S. (2023). Cold Start Latency in Serverless Computing: Current Trends and Mitigation Techniques. Eduzone Journal, 12(2), 134-145.
Golec, M., Walia, G. K., Kumar, M., Cuadrado, F., Gill, S. S., & Uhlig, S. (2023). Cold Start Latency in Serverless Computing: A Systematic Review, Taxonomy, and Future Directions. Journal of the ACM, 37(4), Article 111.
Poccia, D. (2021). AWS Lambda Functions Powered by AWS Graviton2 Processor. https://aws.amazon.com/blogs/aws/aws-lambda-functions-powered-by-aws-graviton2-processor-run-your-functions-on-arm-and-get-up-to-34-better-price-performance (Accessed 24 March 2025).
Kieselhorst, D, Schellhorn M. Serverless Architecture Conference (2024). Serverless Functions with GraalVM on AWS Lambda. https://serverless-architecture.io/blog/serverless_functions_with_graalvm_on_aws_lambda (Accessed 24 March 2025).
Hebbar, R., Chandran, M., and Sudevan, S. (2021). Enhancing Performance of Cloud-based Software Applications with GraalVM and Quarkus. Journal of Cloud Computing Advances, Systems and Applications, 10(1), pp. 1-15.
Minic, B. (2021). Improving Cold Start Times of Java AWS Lambda Functions Using GraalVM. https://shinesolutions.com/blog/aws-lambda-graalvm-cold-start (Accessed 24 March 2025).
Smith, C. (2021). Optimizing Java Applications for Serverless Deployments. International Journal of Advanced Software Engineering, 25(3), pp. 155-170.
Pérez, D., Gómez, J., and Silva, M. (2023). Performance Best Practices Using Java and AWS Lambda. International Journal of Cloud Computing, 12(2), pp. 110-125.
Wang, S. (2021). Thin Serverless Functions with GraalVM Native Image. Serverless Computing Advances, 5(2), pp. 99-110.
Rao, R. (2024). AWS Lambda Cold Starts Explained: What They Are & How to Mitigate Them. Cloud Builder's Journal. https://www.ranthebuilder.cloud/post/is-aws-lambda-cold-start-still-an-issue-in-2024 (Accessed 24 March 2025).
Article Statistics
Copyright License
Copyright (c) 2025 Maksimov Viacheslav Yurievich
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors retain the copyright of their manuscripts, and all Open Access articles are disseminated under the terms of the Creative Commons Attribution License 4.0 (CC-BY), which licenses unrestricted use, distribution, and reproduction in any medium, provided that the original work is appropriately cited. The use of general descriptive names, trade names, trademarks, and so forth in this publication, even if not specifically identified, does not imply that these names are not protected by the relevant laws and regulations.