https://tsvinc.dev/tags/benchmark/ Recent content in Benchmark on tsvinc Hugo en-us Sat, 20 Jun 2026 00:00:00 +0000 https://tsvinc.dev/posts/reactor-removal-benchmark/ Sat, 20 Jun 2026 00:00:00 +0000 https://tsvinc.dev/posts/reactor-removal-benchmark/ <p>We took the entire reactive stack out. R2DBC, Mono, Flux, Reactor Core, loom-carrier, <code>flatMap</code> with its carefully-tuned concurrency arguments. Replaced it with JDBC, HikariCP, and blocking method calls. Same endpoints. Same Netty HTTP server. Same PostgreSQL. Same k6 load profile: 10 to 200 VUs, 500 req/s constant arrival rate, seven scenarios each.</p> <p>Throughput didn&rsquo;t move. Latency dropped. Memory dropped. GC pauses dropped 63% at 256 MB.</p> <h2 id="the-scoreboard">The Scoreboard</h2> <p>This is <a href="https://tsvinc.dev/series/micronaut-256mb/">part of the series</a> benchmarking a Micronaut 5.0.0 app under memory constraints. The reactive version is the one from <a href="https://tsvinc.dev/posts/flatmap-backpressure/">Part 10</a> and <a href="https://tsvinc.dev/posts/loom-carrier-reactive/">Part 11</a> — fully tuned, <code>flatMap</code> concurrency bounded, R2DBC pool sized to match, loom-carrier enabled for virtual threads. The imperative version: JDBC/HikariCP, <code>THREAD_PER_TASK</code> executor for virtual threads, same pool size.</p>