danielfengyu/awesome-jvm

A curated list of awesome loosely low level, performance related JVM stuff. Inspired by awesome-python.

Awesome JVM

A curated list of awesome JVM low level and performance related stuff.

• Awesome JVM
  • Bytecode
  • Garbage collectors
  • Load tools
  • Languages
  • Memory and Concurrency
  • Metaprogramming
  • Native
  • Network
  • Nix tools
  • Profilers
  • Runtimes
  • Virtual Machines
• Resources
  • Communities
  • Documentation
  • Media
  • People
• Contributing

Bytecode

Tools for bytecode manipulation and analysis.

• asmtools - Used to develop tools for the production of Java .class files.
• Byte Buddy - Code generation library creating Java classes at runtime without the help of a compiler.
• Jitescript - Bytecode generation library similar to BiteScript.

Garbage collectors

Garbage collectors for the JVM.

• Azul Pauseless Garbage Collection - Providing continuous, pauseless operation for Java applications.
• Balanced GC - GC policy available in the Java Virtual Machine for IBM WebSphere Application Server V8.
• Epsilon GC - Completely passive GC implementation with bounded allocation limit, and lowest runtime performance overhead possible.
• G1 - The Garbage-First Garbage Collector.
• Shenandoah - Ultra-Low-Pause-Time Garbage Collector.
• The Garbage Collection Handbook - Book that addresses new challenges to garbage collection made by recent advances in hardware and software.

Load tools

Tools that generate load and measure the system accurately without coordinated omission

• Gatling - Asynchronous non-blocking scenario driven load testing tool for testing HTTP servers.
• wrk2 - A constant throughput, correct latency recording variant of wrk.

Languages

Languages running on the JVM.

• Ceylon - Object-oriented, strong and static programming language with an emphasis on immutability, created by Red Hat.
• Clojure - Dialect of Lisp created by Rich Hickey. Dynamically typed with emphasis on functional programming.
• Erjang - A JVM-based Erlang VM.
• Eta - Pure, lazy, strongly typed functional programming language on the JVM.
• Frege - Pure functional programming language in the spirit of Haskell.
• gojava - Java bindings for Go packages.
• Golo - A simple dynamic language that makes extensive usage of invokedynamic.
• Groovy - Optionally typed and dynamic language, with static-typing and static compilation capabilities.
• Java - General-purpose, concurrent, strongly typed, class-based object-oriented language.
• JRuby - Implementation of the Ruby language on the JVM.
• JPHP - PHP on the Java VM.
• Jython - Python for the Java Platform.
• Kawa - Extension of the Scheme language, which is in the Lisp family of programming languages.
• Kotlin - Statically typed programming language for the JVM, Android and the browser.
• LuaJ - Java-centric implementation of lua vm built to leverage standard Java features.
• Nashorn - Lightweight high-performance JavaScript runtime in Java with a native JVM.
• Nodyn - Node.js compatible framework, running on the JVM powered by the DynJS Javascript runtime.
• OCaml-Java - Supports OCaml language v4. Generates plain Java bytecode and have seamless integration with Java.
• Rembulan - Rembulan is an implementation of Lua 5.3 for the JVM, written in pure Java with minimal dependencies.
• Renjin - JVM-based interpreter for the R language for the statistical analysis
• Scala - Strong and static programming language that combine object-oriented and functional programming ideas.
• Xtend - Flexible and expressive dialect of Java, which compiles into Java 5 source code.

Memory and concurrency

Tools and data structures for efficient memory layout and concurrent access.

• Agera - Reactive Programming for Android by Google.
• Agrona - Library of data structures and utility methods that are a common need when building high-performance applications.
• Apache Arrow - A high-performance cross-system data layer for columnar in-memory analytics.
• bloofi - Java implementation of multidimensional Bloom filters
• Cap’n Proto - Insanely fast data interchange format and capability-based RPC system.
• caffeine - A high performance caching library for Java 8.
• Chronicle-Bytes - Low level memory access wrappers.
• Chronicle-Queue - Micro second messaging that stores everything to disk.
• Chronicle-Map - In-memory key-value store designed for low-latency and/or multi-process applications.
• clj-ds - Clojure's data structures modified for use outside of Clojure.
• colfer - Binary serialization format and class generator.
• commons-math - Library of lightweight, self-contained mathematics and statistics components.
• CuckooFilter4J - Bloom filter replacement for approximated set-membership queries.
• cyclops - Integration modules for RxJava, Reactor, FunctionalJava, Guava & Javaslang.
• java-concurrent-hash-trie-map - Java port of a concurrent trie hash map implementation from Scala collections.
• java-hll - Java library for the HyperLogLog algorithm.
• JavaFastPFOR - Library to compress and uncompress arrays of integers very fast.
• fasttuple - Collections that are laid out adjacently in both on- and off-heap memory.
• FlatBuffers - Efficient cross platform serialization library for C++, C#, Go, Java, JavaScript, PHP, and Python.
• geohash - Java utility methods for geohashing.
• gs-collections - Goldman Sachs collections framework.
• hollow - Java library and comprehensive toolset for harnessing small to moderately sized in-memory datasets.
• high-scale-lib - Cliff Click's High Scale Library.
• hppc - High Performance Primitive Collections.
• injector - A new Executor for Java.
• java-string-similarity - String similarity and distance measures, including Levenshtein edit distance and sibblings, Jaro-Winkler, Longest Common Subsequence, cosine similarity etc.
• JCTools - Concurrent data structures currently missing from the JDK.
• Koloboke - Java Collections til the last breadcrumb of memory and performance.
• LevelDB - Rewrite (port) of LevelDB in Java.
• lightweight_trie - A very memory-efficient trie (radix tree) implementation.
• lmdbjni - Java API to LMDB (HawtJNI) which is an ultra-fast, ultra-compact key-value embedded data store written in C.
• lmdbjava - Java API to LMDB (JNR) which is an ultra-fast, ultra-compact key-value embedded data store written in C.
• low-gc-membuffers - In-memory circular buffers that use direct ByteBuffers to minimize GC overhead.
• lwjgl3 - Java library that enables cross-platform access to popular native APIs useful in the development of graphics (OpenGL), audio (OpenAL) and parallel computing (OpenCL) applications.
• MapDB - Collections backed by off-heap or on-disk storage.
• netty-buffers - Memory buffer pool implementation similar to jemalloc.
• ObjectLayout - A layout-optimized Java data structure package.
• ohc - Java large off heap cache developed for Apache Cassandra 3.0.
• okio - Modern Java IO library that do clever things to save CPU and memory.
• one-nio - library for building high performance Java servers.
• onyx-java - Mirrors the Onyx Platform core API by providing a Java equivalent for each component of an Onyx workflow.
• parquet - Columnar storage format that uses the record shredding and assembly algorithm described in the Dremel paper.
• PauselessHashMap - A java.util.HashMap compatible map that won't stall puts or gets when resizing.
• pcollections - A Persistent Java Collections Library.
• protobuf - Google's data interchange format.
• Quasar - Lightweight threads and actors for the JVM.
• rtree - Immutable in-memory R-tree and R*-tree implementations in Java with reactive api.
• Reactive Streams - Standard for asynchronous stream processing with non-blocking back pressure.
• RoaringBitmap - A better compressed bitset in Java.
• rollinghashjava - Rolling hash functions in Java.
• Reactor - Reactive data applications on the JVM for Java, Groovy, Clojure and other.
• RxJava - Library for composing asynchronous and event-based programs using observable sequences.
• SmoothieMap - java.util.Map impl with worst put latencies more than 100 times smaller than java.util.HashMap.
• Simple Binary Encoding - High Performance Message Codec.
• DataSketches - A Java software library of stochastic streaming algorithms.
• stormpot - A fast object pool for the JVM.
• stream-lib - A Java library for summarizing data in streams for which it is infeasible to store all events.
• TraneIO - High-performance implementation of the Future abstraction.
• transducers-java - Composable algorithmic transformations independent from the context of their input and output sources.
• vavr - Functional Library for Java 8+.
• Zero-Allocation-Hashing - Hashing any sequences of bytes in Java, including all kinds of primitive arrays, buffers, CharSequences and more.

Metaprogramming

Parsers, interpreters, compilers and source generation targeted for the JVM.

• Antlr - Parser generator for reading, processing, executing, or translating structured text or binary files.
• auto - A collection of source code generators for Java.
• Apache Calcite - Dynamic data management framework and SQL parser plugin.
• Checker Framework - Compiler plug-ins that find bugs or verify their absence.
• compile-testing - Testing tools for javac and annotation processors.
• derive4j - Algebraic data types constructors, pattern-matching, morphisms, optics and typeclasses.
• error-prone - Catch common Java mistakes as compile-time errors.
• GHCVM - A Haskell to JVM compiler that supports GHC Haskell.
• Graal - New experimental just-in-time compiler for Java that is integrated with the HotSpot virtual machine.
• immutables - Generate simple, safe and consistent value objects.
• javacc - Parser generator for use with Java.
• javaparser - Java 1.8 Parser and Abstract Syntax Tree for Java.
• JavaPoet - A Java API for generating .java source files.
• jparsec - Builds mini parsers in pure Java a la Haskell Parsec.
• JSweet - A transpiler from Java to TypeScript/JavaScript.
• MPS - Design and build extensible DSLs and editors.
• parboiled - Parsing of arbitrary input text based on parsing expression grammars.
• Sulong - LLVM IR interpreter written in Java using Truffle and Graal.
• TeaVM - Ahead-of-time translating compiler (transpiler) from Java bytecode to JavaScript.
• Truffle - Framework for implementing languages as simple interpreters.
• Xtext - Framework for development of programming languages and DSLs.

Native

Interconnecting JVM and native code

• hawtjni - A JNI code generator based on the JNI generator used in Eclipse SWT.
• j2v8 - Java API for Google's V8 JavaScript engine.
• JavaCPP - JavaCPP provides efficient access to native C++ inside Java.
• jnr-ffi - Load native libraries without writing JNI code by hand.
• jssembly - Execution of native assembly from Java.
• NuProcess - A low-overhead, non-blocking I/O, external Process execution implementation for Java.
• Project Panama - Enriching the connections between the JVM and APIs used by C programmers.

Network

Tools for network programming, packet capture, monitoring, testing and resiliency.

• Aeron - Efficient reliable UDP unicast, UDP multicast, and IPC message transport.
• armeria - Asynchronous RPC/API client/server library built on top of Java 8, Netty 4.1, HTTP/2, and Thrift.
• comcast - Simulating shitty network connections.
• gor - HTTP traffic replay in real-time.
• gRPC - A high performance, open source, general RPC framework that puts mobile and HTTP/2 first.
• jRT - Measures response time of a java application to socket-based requests.
• K3PO - Create arbitrary network traffic and behavior to certify whether a network endpoint behaves correctly.
• muxy - Simulating real-world distributed system failures.
• Netty - Async event-driven network library for high performance protocol servers & clients.
• okhttp - An HTTP+HTTP/2 client for Android and Java applications.
• RSocket - RSocket is a binary protocol for use on byte stream transports such as TCP, WebSockets, and Aeron.
• SimianArmy - Resiliency tool that helps ensure that your applications can tolerate random instance failures.
• pcap4j - Java library for capturing, crafting, and sending packets using libpcap.
• pig - A Linux packet crafting tool.
• tcpdump - Packet analyzer for network traffic capture.
• tcpflow - Captures TCP connections flows in a way that is convenient for protocol analysis and debugging.
• tcpreplay - Pcap editing and replay tools.

Nix tools

*Useful nix tools when profiling the JVM and interaction with the host environment

• atoptool - Logging of system and process activity for long-term analysis, highlighting overloaded system.
• Flame Graphs - Visualization of profiled software, allowing the most frequent code-paths to be identified quickly and accurately.
• javap - Disassembles class files into code that reflects the java bytecode.
• jhat - Java Heap Analysis Tool
• jhsdb - Launch a postmortem debugger to analyze the content of a core-dump from a crashed JVM.
• jinfo - Prints configuration information for a given process.
• jstack - Prints stack traces of threads for a given Java process.
• jstat - Monitors GC and compiler statistics in the JVM.
• hwloc - Reports the structure of the processor, number of cores, hyperthreads and cache size.
• likwid - Read hardware performance counters on Intel and AMD processors.
• numactl - Control NUMA policy for processes or shared memory.
• oprofile - System-wide hardware performance monitoring with easy-to-use interface at low overhead.
• perf - Linux profiling with performance counters.
• perf-tools - Performance analysis tools based on Linux perf_events (aka perf) and ftrace.
• sysdig - Capture system state and activity from a running Linux instance, then save, filter and analyze.
• sysstat - Performance monitoring tools for Linux.
• taskset - Retrieve or set a processes’s CPU affinity.
• tiptop - Like top but also shows instructions per cycle (IPC).

Profilers

Tools that provide profiling and tracing information to aid program optimization

• allocation-instrumenter - Java agent that rewrites bytecode to instrument allocation sites.
• aprof - Java memory allocation profiler.
• BTrace - a safe, dynamic tracing tool for the Java platform.
• Chronon - Record your entire java program. Replay on any machine.
• GCViewer - GCViewer is a tool that visualizes verbose GC output.
• hawkshaw - Tools for tracking down memory / JVM problems & generating predictable-as-possible VM behaviour.
• HdrHistogram - A Histogram that supports recording and analyzing sampled data value counts.
• hdrhistogram-metrics-reservoir - A Metrics Reservoir implementation backed by HdrHistogram.
• HdrLogProcessing - Utilities for HDR Histogram logs manipulation.
• heapster - Production heap profiling for the JVM.
• honest-profiler - Sampling JVM profiler without the safepoint sample bias.
• jamm - Measure actual object memory use including JVM overhead.
• Java Flight Recorder (JFR) - Tool for collecting diagnostic and profiling data about a running Java application with almost no performance overhead.
• java-sizeof - Memory consumption estimator for Java.
• jcstress - Experimental harness and tests to aid the research in the correctness of concurrency support in the JVM, class libraries, and hardware.
• jfr-flame-graph - Converting JFR Method Profiling Samples to FlameGraph compatible format.
• jitwatch - Log analyser / visualiser for Java HotSpot JIT compiler.
• jitwatch-intellij - JITWatch plugin for IntelliJ IDEA.
• jHiccup - jHiccup is an open source tool designed to measure the pauses and stalls associated with an application’s underlying Java runtime platform.
• jmh - Micro benchmarks written in Java and other languages targetting the JVM.
• jmh-compare-gui - GUI for comparing JMH results.
• JOL - Analyze actual object layout schemes, footprint, and references in JVMs.
• JProfiler - Helps resolve performance bottlenecks, pin down memory leaks and understand threading issues.
• JVMTI - Provide a native API to inspect the state and to control the execution of applications running in the JVM.
• jvmtop - Lightweight console application to monitor running jvms on a machine in top-like manner.
• MAT - Java heap analyzer that help find memory leaks and reduce memory consumption.
• leakcanary - A memory leak detection library for Android and Java.
• metrics - Measure the behavior of critical components in production environment.
• Overseer - Low-Level Hardware Monitoring and Management for Java.
• perf-map-agent - Generate method mappings to use with the linux perf tool.
• perfj - Linux perf for java programs.
• polarbear - A tool to help diagnose OutOfMemoryError conditions.
• Riemann JVM Profiler - JVM agent which sends function-level profiler telemetry to a Riemann server for analysis, visualization, and storage.
• statsd-jvm-profiler - JVM agent profiler that sends profiling data to StatsD.
• Swiss Java Knife - Small set of tools for JVM troublshooting, monitoring and profiling.
• Takipi - Tells you when and why code breaks in production.
• Tracer - Manages custom trace identifiers and carries them through distributed systems.
• YourKit - Fully featured, easy to use, low overhead profiler.
• Zipkin - A distributed tracing system gather timing data for disparate services developed by Twitter.

Runtimes

Tools for managing jvm runtime processes

• Capsule - Dead-Simple Packaging and Deployment for JVM Apps.
• CRaSH - The shell for the Java Platform.
• Drip - Fast JVM launching without the hassle of persistent JVMs.
• HotswapAgent - Redefine classes at runtime and skip the redeploy process.
• jvmkill - Agent that forcibly terminates the JVM when it is unable to allocate memory or create a thread.
• Nailgun - Nailgun is a client, protocol, and server for running Java programs from the command line without incurring the JVM startup overhead.

Virtual Machines

Virtual machines that implement the JVM specification or parts of it.

• Avian - Lightweight highly portable JVM with an option for AOT compilation.
• Dalvik - Android runtime (ART) is the managed runtime used by applications and some system services on Android.
• DCEVM - Modification of Java HotSwap VM with unlimited support for reloading classes at runtime.
• HotSpot - HotSpot virtual machine maintained and distributed by Oracle Corporation.
• IBM J9 - JVM developed by IBM.
• J2ObjC - Translator from Java source to Objective-C code. Keeps shared code between iOS native apps and Android native apps.
• jvm.go - A JVM written in Go.
• ParparVM - An Open Source Java bytecode to C translator for iOS native development. Designed as a part of the Codename One WORA for mobile project.
• RoboVM - Create native iOS and Android apps in Java. - Discontinued
• Zing - The only JVM that eliminates Java garbage collection pauses for large heap sizes.
• Zulu - The only certified multi-platform build of OpenJDK: Free, 100% open source Java.

Resources

Documentation

Documentation related to JVM

• JVM Anatomy Park - mini-post series where every post goes deep for only a single topic by Aleksey Shipilёv.
• Coordinated Omission problem - Discussion on Mechanical Sympathy.
• False sharing - Threads impact the performance of each other while modifying independent variables sharing the same cache line. Martin Thompson.
• The JVM specification - The Java Virtual Machine Specification Java SE 8 Edition.
• The Java Memory Model - Starting point for discussions of and information concerning the Java Memory Model.
• The JSR-133 Cookbook for Compiler Writers - Unofficial guide to implementing the new Java Memory Model (JMM) specified by JSR-133.
• Garbage Collection Tuning Guide - HotSpot Virtual Machine Garbage Collection Tuning Guide.
• Safepoints - Where is my safepoint? Nitsan Wakart.
• Topics in High-Performance Messaging - Design decisions, experience and constraints explained in high performance messaging systems.
• Top 10 Performance Mistakes - Digest of the top 10 performance related mistakes Martin Thompson has seen in production.
• The USE method - The Utilization Saturation and Errors (USE) Method is a methodology for analyzing the performance of any system. Brendan Gregg.
• An introduction to distributed systems - Kyle Kingsbury (author of Jepsen).
• Using JDK 9 Memory Order Modes - For expert programmers familiar with Java concurrency, but unfamiliar with the memory order modes available in JDK 9 provided by VarHandles.
• CPU Utilization is Wrong - Measure instructions per cycle (IPC) for CPU utilization. Brendan Gregg.

Communities

Active discussions.

• concurrency-interest - Discussion list for JSR-166.
• hotspot-compiler-dev - Technical discussion about the development of the HotSpot bytecode compilers.
• hotspot-dev - HotSpot development mailing list.
• hotspot-gc-dev - Technical discussion about the development of the HotSpot garbage collectors.
• mechanical-sympathy - Discussing how to code sympathetically to and measure the underlying stack/platform so good performance can be extracted.
• Performance Java User's Group - For expert Java developers who want to push their systems to the next level
• Virtual Machine Meetup - Venue for discussing the latest research and developments in the area of managed language execution.

Media

Videos, podcasts and other media related to JVMs

• G1 Garbage Collector in Java 8/9 - Kirk Pepperdine.
• Extreme Profiling: Digging Into Hotspots - Nitsan Wakart.
• Java vs. C Performance - Cliff Click.
• Why JNI is slow? - Cliff Click
• A Crash Course in Modern Hardware - Cliff Click
• Java Profiling from the Ground Up - Nitsan Wakart.
• The Illusion of Execution - Nitsan Wakart.
• Mythbusting Modern Hardware to Gain 'Mechanical Sympathy' - Martin Thompson.
• Designing for Performance - Martin Thompson.
• How NOT to Measure Latency - Gil Tene.
• JVM Language Summit 2015 - JVM Language Summit 2015.
• JVM Language Summit 2016 - JVM Language Summit 2016.
• Bits of advice for VM writers - Cliff Click.
• Understanding Java garbage collection ... - Gil Tene.
• Faster Object Arrays - Gil Tene at GOTO Conferences.
• Java Memory Model Pragmatics - Aleksey Shipilev.
• With GC Solved, What Else Makes a JVM Pause? - John Cuthbertson.
• JVM Mechanics - Douglas Hawkins.
• Give me 15 minutes and I'll change your view of Linux tracing - Brendan Gregg.
• Analyzing and Debugging the Java HotSpot VM at the OS Level - Volker Simonis.

People

People that influence JVM development and/or the community around it

• Aleksey Shipilëv - Developing Oracle/Open JDK/Hotspot and other Java-related technologies.
• Brian Goetz - Java Language Architect at Oracle.
• Ben Christensen - Facebook, Netflix, Apple engineering.
• Brendan Gregg - Cloud performance, kernel engineer, speaker, author.
• Charles Nutter - JRuby guy.
• Cliff Click - Creator of the HotSpot Server Compiler.
• Dave Dice - Senior research scientist in the Scalable Synchronization Research Group within Oracle.
• Dávid Karnok - RxJava committer that blogs about advanced RxJava.
• Doug Lea - Author of the Java memory model.
• Gil Tene - Azul Systems.
• John Rose - HotSpot developer.
• Jonas Bonér - Founder & CTO of Lightbend.
• Marcus Lagergren - Java language team alumnus.
• Mark Reinhold - Chief Architect, Java Platform Group, Oracle.
• Martin Thompson - Pasty faced performance gangster.
• Martijn Verburg - Java Champion.
• Kirk Pepperdine - Working in high performance and distributed computing for nearly 20 years.
• Nitsan Wakart - Azul Systems.
• Norman Maurer - Netty developer.
• Paul Phillips - Forever undisputed SLOC Scala compiler dev.
• Peter Lawrey - Innovative developer of high performance Java systems for competitive advantage.
• Richard Warburton - Developer, Speaker, Author.
• Todd L. Montgomery - Ex-CTO, Ex-NASA researcher, network geek, messaging middleware designer.
• Stéphane Maldini - Project Reactor Lead @Pivotal.
• Stuart Marks - Doctor Deprecator. Java/JDK/OpenJDK developer
• Viktor Klang - Deputy CTO at Typesafe Inc.

Contributing

Contributions are very welcome!

Please have a look at contributing.md for guidelines.