Performance and capacity planning in Java based data grids
summary
Project Radargun is an open source Java benchmarking framework that automates coordinated multi‑node runs with a master and workers, performing validation, warm‑up, pristine runs and centralized reporting. It runs plug‑ins for Infinispan, JBoss Cache, Hazelcast, Terracotta and others and ships web session and TPC‑C benchmarks to exercise low‑contention and transactional workloads. Output shows TPS per node and memory/CPU/GC charts across cluster sizes (examples run 2–4 nodes), highlights uneven key distribution, and catches regressions in throughput and resource use.
AI-written summary · about
chapters (8)
- 0:00Introduction and speaker background
- 1:11What are in-memory data grids
- 6:23Why benchmarks are necessary
- 7:52How Radar Gun works
- 9:38Radar Gun configuration example
- 12:15Default benchmarks and reports
- 14:10Custom benchmarks and plugins
- 16:44Summary and Q&A
related talks
01h 23mBuilding Highly-Scalable Spring Applications with In-Memory, Distributed Data Grids
54mThe JBoss Data Grid, or Enterprisegrade Infinispan
38mBoost Your Applications with Distributed Caches/Datagrid
05mStartup Showcase: Hazelcast
47mBoosting your applications with distributed caches/datagrids
43mCaching and Beyond - Smarter Data Processing with Java
44mBoosting your applications with distributed caches/datagrids
01h 01mWeb Clustering, Integration with Terracotta, BigMemory, Quartz & Grails