Tools
Tools
Pandita, Vishesh (2024) Benchmarking the Performance of Java Virtual Threads in High-Throughput Workloads. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (1MB) | Preview |
Preview |
PDF (Configuration Manual)
Download (2MB) | Preview |
Abstract
This research evaluates the performance of Java Virtual Threads against traditional platform threads in high-throughput workloads and benchmarks the process. Java Virtual Threads were introduced as part of Project Loom in JDK 21 to enable optimal hardware utilization for applications written in a thread-per-request model. Virtual Threads, introduced in JDK 21 in 2023, required in-depth testing to evaluate their contributions to Java’s concurrent and asynchronous programming. In this research, attention was given to creating a fair test environment that will not influence the results of benchmarking Java applications under high workloads. The benchmarking was done on both CPU and I/O intensive workloads to test and evaluate the performance of virtual threads compared to platform threads.
In CPU-intensive workloads, Java applications with traditional and virtual threads performed similarly. However, when it comes to I/O intensive workloads, Java virtual thread applications perform better than Java platform thread applications. Applications using Java virtual threads used less memory and had better latency compared to traditional platform threads. Throughput was increased by 60.79% while Latency was decreased by 28.8% while comparing virtual threads with platform threads. Memory usage and CPU utilization also saw an improvement of 36.36% and 14.29% respectively.
Both virtual and platform threads have their strengths and weaknesses. It can not be said that virtual threads are better than platform threads. However, it highly depends on the work scenarios and which metrics one needs to prioritize. Virtual threads performed better in blocking operations but Platform threads are more stable.
| Item Type: | Thesis (Masters) |
|---|---|
| Supervisors: | Name Email Samarawickrama, Yasantha UNSPECIFIED |
| Subjects: | Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Cloud computing |
| Divisions: | School of Computing > Master of Science in Cloud Computing |
| Depositing User: | Ciara O'Brien |
| Date Deposited: | 16 Jul 2025 09:53 |
| Last Modified: | 16 Jul 2025 09:53 |
| URI: | https://norma.ncirl.ie/id/eprint/8134 |
Actions (login required)
 |
View Item |