https://blog.turboawesome.win/tags/jvm/ Recent content in Jvm on Bits, Trades & Systems Hugo en-us Thu, 24 May 2018 09:38:00 +0000 https://blog.turboawesome.win/2018/05/project-loom-preview-virtual-threads-and-what-they-mean-for-server-code/ Thu, 24 May 2018 09:38:00 +0000 https://blog.turboawesome.win/2018/05/project-loom-preview-virtual-threads-and-what-they-mean-for-server-code/ Project Loom's virtual threads promise to fix Java's thread-per-request scalability problem by making millions of lightweight threads practical. Here's what the early design looked like and why it matters for server-side Java. https://blog.turboawesome.win/2017/11/from-java-8-to-java-11-in-a-regulated-environment-what-actually-broke/ Wed, 08 Nov 2017 11:23:00 +0000 https://blog.turboawesome.win/2017/11/from-java-8-to-java-11-in-a-regulated-environment-what-actually-broke/ Migrating a large financial institution's Java estate from Java 8 to Java 11 involved removing internal APIs, JPMS module system friction, and more dependency surprises than expected. What actually broke and how we fixed it. https://blog.turboawesome.win/2017/04/reading-gc-logs-like-a-detective/ Tue, 18 Apr 2017 09:11:00 +0000 https://blog.turboawesome.win/2017/04/reading-gc-logs-like-a-detective/ GC logs are the most information-dense diagnostic output the JVM produces, and most engineers never learn to read them properly. A guide to interpreting G1GC logs and using them to diagnose real production problems. https://blog.turboawesome.win/2016/08/heap-dumps-and-flight-recorder-diagnosing-jvm-memory-problems-in-production/ Wed, 24 Aug 2016 14:55:00 +0000 https://blog.turboawesome.win/2016/08/heap-dumps-and-flight-recorder-diagnosing-jvm-memory-problems-in-production/ Heap dumps and Java Flight Recorder are the two tools that diagnose production JVM memory problems. Understanding how to capture them without killing the service, and how to read what they tell you, is a gap in most JVM engineers' toolkits. https://blog.turboawesome.win/2016/04/why-the-risk-team-chose-clojure-and-why-it-made-sense/ Tue, 05 Apr 2016 14:17:00 +0000 https://blog.turboawesome.win/2016/04/why-the-risk-team-chose-clojure-and-why-it-made-sense/ A Lisp in a large financial institution sounds like a punchline. The reasoning behind the decision was more principled than the stereotype suggests. https://blog.turboawesome.win/2015/11/five-years-in-high-frequency-trading-what-i-actually-learned/ Thu, 12 Nov 2015 16:03:00 +0000 https://blog.turboawesome.win/2015/11/five-years-in-high-frequency-trading-what-i-actually-learned/ Looking back at five years of low-latency systems work: what the environment teaches you that you can't learn elsewhere, and what habits it instils that don't always transfer cleanly to other domains. https://blog.turboawesome.win/2015/10/zgc-and-shenandoah-what-low-pause-gc-means-for-trading-systems/ Thu, 01 Oct 2015 09:44:00 +0000 https://blog.turboawesome.win/2015/10/zgc-and-shenandoah-what-low-pause-gc-means-for-trading-systems/ In 2015 the JVM GC landscape was about to change significantly. ZGC and Shenandoah were in development, promising sub-millisecond GC pauses regardless of heap size. Here's what the previews looked like and what they meant for latency-sensitive Java. https://blog.turboawesome.win/2015/08/scala-akka-actors-for-trading-workflows-promises-and-pitfalls/ Wed, 19 Aug 2015 11:28:00 +0000 https://blog.turboawesome.win/2015/08/scala-akka-actors-for-trading-workflows-promises-and-pitfalls/ We evaluated Akka actors for a workflow orchestration layer in the trading system. The actor model's isolation and message-passing properties were genuinely useful — and its failure modes were genuinely painful. https://blog.turboawesome.win/2015/05/understanding-safepoints-the-jvm-pauses-nobody-talks-about/ Wed, 27 May 2015 14:38:00 +0000 https://blog.turboawesome.win/2015/05/understanding-safepoints-the-jvm-pauses-nobody-talks-about/ GC pauses get all the attention, but safepoints are a broader category of JVM stop-the-world events that affect latency in ways GC tuning can't fix. https://blog.turboawesome.win/2014/08/choosing-a-gc-collector-for-low-latency-java-a-practical-comparison/ Wed, 06 Aug 2014 10:18:00 +0000 https://blog.turboawesome.win/2014/08/choosing-a-gc-collector-for-low-latency-java-a-practical-comparison/ CMS vs G1 vs parallel GC vs no-GC strategies. The tradeoffs as they actually played out in a latency-sensitive financial application, with real numbers. https://blog.turboawesome.win/2014/04/off-heap-memory-in-java-sun.misc.unsafe-and-chronicle-map/ Wed, 02 Apr 2014 10:17:00 +0000 https://blog.turboawesome.win/2014/04/off-heap-memory-in-java-sun.misc.unsafe-and-chronicle-map/ The Java heap is GC'd. Off-heap memory is not. For large data structures that don't benefit from GC management and do benefit from predictable latency, off-heap allocation via Unsafe and Chronicle Map is the right tool — if you respect its dangers. https://blog.turboawesome.win/2014/01/scala-on-the-hot-path-where-the-abstraction-cost-goes/ Wed, 08 Jan 2014 09:44:00 +0000 https://blog.turboawesome.win/2014/01/scala-on-the-hot-path-where-the-abstraction-cost-goes/ Scala's abstractions are elegant — but on the JVM, elegance has a bytecode cost. Understanding where Scala generates boxing, closures, and virtual dispatch tells you which patterns are safe in hot paths and which aren't. https://blog.turboawesome.win/2013/11/why-average-latency-is-a-lie-hdrhistogram-and-measuring-what-matters/ Wed, 27 Nov 2013 11:14:00 +0000 https://blog.turboawesome.win/2013/11/why-average-latency-is-a-lie-hdrhistogram-and-measuring-what-matters/ Average latency hides the distribution. HdrHistogram gives you a complete picture of latency across the full range — and it does so without distorting the system you're measuring. https://blog.turboawesome.win/2013/07/jvm-jit-compilation-what-the-c2-compiler-does-to-your-loops/ Tue, 30 Jul 2013 14:11:00 +0000 https://blog.turboawesome.win/2013/07/jvm-jit-compilation-what-the-c2-compiler-does-to-your-loops/ The HotSpot C2 compiler can transform your Java bytecode into surprisingly efficient native code — but only under specific conditions. Understanding those conditions explains a class of mysterious performance cliffs. https://blog.turboawesome.win/2012/11/stop-the-world-gc-pauses-killed-our-sla-and-what-we-did-about-it/ Tue, 13 Nov 2012 11:08:00 +0000 https://blog.turboawesome.win/2012/11/stop-the-world-gc-pauses-killed-our-sla-and-what-we-did-about-it/ How garbage collection pauses blew our latency SLA during peak market hours, and the three-phase approach we used to get it under control. https://blog.turboawesome.win/2012/06/why-your-java-app-is-slow-before-it-even-starts-classloading-deep-dive/ Mon, 18 Jun 2012 09:11:00 +0000 https://blog.turboawesome.win/2012/06/why-your-java-app-is-slow-before-it-even-starts-classloading-deep-dive/ Java classloading happens lazily, on demand, and under lock. Understanding the classloader hierarchy and what triggers loading explains a category of JVM startup and first-call latency problems.