10/26/2017 Universal Java GC analysis tool - Java Garbage collection log analysis made easy

Similar documents
It s Good to Have (JVM) Options

G1 Garbage Collector Details and Tuning. Simone Bordet

HBase Practice At Xiaomi.

Lesson 2 Dissecting Memory Problems

Java & Coherence Simon Cook - Sales Consultant, FMW for Financial Services

Java Performance Tuning

The Garbage-First Garbage Collector

Java Memory Management. Märt Bakhoff Java Fundamentals

Do Your GC Logs Speak To You

JVM Memory Model and GC

Java Application Performance Tuning for AMD EPYC Processors

NG2C: Pretenuring Garbage Collection with Dynamic Generations for HotSpot Big Data Applications

2011 Oracle Corporation and Affiliates. Do not re-distribute!

The Fundamentals of JVM Tuning

Garbage collection. The Old Way. Manual labor. JVM and other platforms. By: Timo Jantunen

Java Performance Tuning From A Garbage Collection Perspective. Nagendra Nagarajayya MDE

Attila Szegedi, Software

Fundamentals of GC Tuning. Charlie Hunt JVM & Performance Junkie

JVM Performance Study Comparing Java HotSpot to Azul Zing Using Red Hat JBoss Data Grid

Insight Case Studies. Tuning the Beloved DB-Engines. Presented By Nithya Koka and Michael Arnold

Unified JVM Logging in JDK 9

Configuring the Heap and Garbage Collector for Real- Time Programming.

Typical Issues with Middleware

Exploiting the Behavior of Generational Garbage Collector

The G1 GC in JDK 9. Erik Duveblad Senior Member of Technical Staf Oracle JVM GC Team October, 2017

JVM Troubleshooting MOOC: Troubleshooting Memory Issues in Java Applications

THE TROUBLE WITH MEMORY

JVM and application bottlenecks troubleshooting

Java Platform, Standard Edition HotSpot Virtual Machine Garbage Collection Tuning Guide. Release 10

How to keep capacity predictions on target and cut CPU usage by 5x

Java Garbage Collection Best Practices For Tuning GC

Generational Garbage Collection Theory and Best Practices

Low latency & Mechanical Sympathy: Issues and solutions

MEMORY MANAGEMENT HEAP, STACK AND GARBAGE COLLECTION

JVM Performance Study Comparing Oracle HotSpot and Azul Zing Using Apache Cassandra

Garbage Collection. Hwansoo Han

OS-caused Long JVM Pauses - Deep Dive and Solutions

Troubleshooting Memory Problems in Java Applications

Monitoring Agent for Tomcat 6.4 Fix Pack 4. Reference IBM

Concurrent Garbage Collection

The Z Garbage Collector Scalable Low-Latency GC in JDK 11

Real Time: Understanding the Trade-offs Between Determinism and Throughput

IBM Cognos ReportNet and the Java Heap

Enabling Java-based VoIP backend platforms through JVM performance tuning

A Study Paper on Performance Degradation due to Excessive Garbage Collection in Java Based Applications using Profiler

Garbage-First Garbage Collection by David Detlefs, Christine Flood, Steve Heller & Tony Printezis. Presented by Edward Raff

Kodewerk. Java Performance Services. The War on Latency. Reducing Dead Time Kirk Pepperdine Principle Kodewerk Ltd.

Latency and Throughput

Workload Characterization and Optimization of TPC-H Queries on Apache Spark

Performance Sentry VM Provider Objects April 11, 2012

The C4 Collector. Or: the Application memory wall will remain until compaction is solved. Gil Tene Balaji Iyengar Michael Wolf

Contents. Created by: Raúl Castillo

High Performance Managed Languages. Martin Thompson

IBM V7000 Unified R1.4.2 Asynchronous Replication Performance Reference Guide

Java Performance Tuning and Optimization Student Guide

Lesson 3 Diagnostic Data Collection and Analysis tools

JVM Memory Management and ColdFusion Log Analysis

Page 2 of 6 SUT Model Form Factor CPU CPU Characteristics Number of Systems 1 Nodes Per System 1 Chips Per System 2 Hardware hw_1 Cores Per System 44

Java Performance: The Definitive Guide

JVM Performance Tuning with respect to Garbage Collection(GC) policies for WebSphere Application Server V6.1 - Part 1

Application Management Webinar. Daniela Field

Welcome to the session...

Part I Garbage Collection Modeling: Progress Report

New Java performance developments: compilation and garbage collection

WHO AM I.

Understanding Java Garbage Collection

Generation scavenging: A non-disruptive high-performance storage reclamation algorithm By David Ungar

Monitoring Agent for Tomcat 6.4 Fix Pack 8. Reference IBM

Efficient Runtime Tracking of Allocation Sites in Java

Large Memory Pages Part 2

Understanding Garbage Collection

Performance Evaluation of WiFiRe using OPNET

Shenandoah: Theory and Practice. Christine Flood Roman Kennke Principal Software Engineers Red Hat

JBoss Performance Tuning. Bill Meyer JBoss Sr. Solutions Architect

Understanding Java Garbage Collection

Monitoring Agent for Microsoft Hyper-V Server Fix Pack 12. Reference IBM

Algorithms for Dynamic Memory Management (236780) Lecture 4. Lecturer: Erez Petrank

A Case for Packing and Indexing in Cloud File Systems

Designing experiments Performing experiments in Java Intel s Manycore Testing Lab

Towards High Performance Processing in Modern Java-based Control Systems. Marek Misiowiec Wojciech Buczak, Mark Buttner CERN ICalepcs 2011

Configuring Active Latency Monitoring

Memory management has always involved tradeoffs between numerous optimization possibilities: Schemes to manage problem fall into roughly two camps

Google File System. Arun Sundaram Operating Systems

Tuning Cognos ReportNet for a High Performance Environment

CSE 120: Principles of Operating Systems. Lecture 10. File Systems. November 6, Prof. Joe Pasquale

A JVM Does What? Eva Andreasson Product Manager, Azul Systems

Understanding Java Garbage Collection

A Side-channel Attack on HotSpot Heap Management. Xiaofeng Wu, Kun Suo, Yong Zhao, Jia Rao The University of Texas at Arlington

Shenandoah An ultra-low pause time Garbage Collector for OpenJDK. Christine H. Flood Roman Kennke

ZENworks 2017 Update 1 ZENworks Diagnostics and Probe Guide. July 2017

Vendavo Tuning Guide

Dynamic Vertical Memory Scalability for OpenJDK Cloud Applications

PowerVault MD3 SSD Cache Overview

@joerg_schad Nightmares of a Container Orchestration System

Product Guide. McAfee Performance Optimizer 2.2.0

Sat 1/6/2018 JAVA (JVM) MEMORY MODEL MEMORY MANAGEMENT IN JAVA... 3 WHERE STORAGE LIVES... 3

CS252 S05. CMSC 411 Computer Systems Architecture Lecture 18 Storage Systems 2. I/O performance measures. I/O performance measures

SDK/RTE for Debian Linux on Intel Itanium Processors Release Notes

High Performance Managed Languages. Martin Thompson

Estimate performance and capacity requirements for InfoPath Forms Services 2010

Transcription:

Analysis Report GC log le: atlassian-jira-gc-2017-10-26_0012.log.0.current Duration: 14 hrs 59 min 51 sec System Time greater than User Time In 25 GC event(s), 'sys' time is greater than 'usr' time. It's not a healthy sign. Timestamp User Time (secs) Sys Time (secs) Real Time (secs) 2017-10-26T01:55:27 0.35 0.71 0.15 2017-10-26T03:01:44 0.62 0.84 0.24 2017-10-26T03:09:26 0.15 0.61 0.11 2017-10-26T03:16:04 0.2 0.44 0.09 2017-10-26T03:20:10 0.38 0.59 0.13 2017-10-26T03:23:20 0.43 0.76 0.16 2017-10-26T04:03:28 0.34 0.79 0.16 2017-10-26T04:06:43 0.24 0.99 0.2 2017-10-26T04:11:08 0.32 1.0 0.29 2017-10-26T04:18:59 0.31 0.8 0.16 2017-10-26T04:26:29 0.58 1.55 0.31 2017-10-26T04:33:42 0.13 0.48 0.15 2017-10-26T04:38:59 0.22 0.72 0.12 2017-10-26T05:00:44 0.43 0.73 0.15 2017-10-26T05:23:55 0.12 0.7 0.14 2017-10-26T05:25:39 0.35 1.2 0.36 2017-10-26T05:33:01 0.26 1.36 0.21 2017-10-26T06:00:44 0.59 1.4 0.26 2017-10-26T06:10:57 0.48 3.84 0.59 2017-10-26T06:31:59 0.21 0.54 0.22 2017-10-26T06:33:03 0.23 1.28 0.21 5 more events... Read our recommendations to reduce sys time (https://blog.gceasy.io/2016/12/11/sys-time-greater-than-user-time/) http://gceasy.io/diamondgc-report.jsp?p=yxjjagl2zwqvmjaxny8xmc8yni8tlwf0bgfzc2lhbi1qaxjhlwdjltiwmtctmtatmjzfmdaxmi5sb2cumc5jdxj 1/9

JVM Heap Size Generation Allocated Peak Young Generation 9.92 gb 9.6 gb Old Generation 6.08 gb 12.7 gb Meta Space 1.03 gb 35.44 mb Young + Old + Meta space 17.03 gb 15.63 gb Key Performance Indicators (Important section of the report. To learn more about KPIs, click here (https://blog.gceasy.io/2016/10/01/garbage-collection-kpi/)) 1 Throughput : 99.584% 2 Latency: Avg GC Time 77 ms Max GC Time 2 sec 870 ms GC Duration Time Range : Duration (secs) No. of GCs Percentage 0-1 2853 99.895% 1-2 2 99.965% 2-3 1 100.0% http://gceasy.io/diamondgc-report.jsp?p=yxjjagl2zwqvmjaxny8xmc8yni8tlwf0bgfzc2lhbi1qaxjhlwdjltiwmtctmtatmjzfmdaxmi5sb2cumc5jdxj 2/9

Interactive Graphs (All graphs are zoomable) http://gceasy.io/diamondgc-report.jsp?p=yxjjagl2zwqvmjaxny8xmc8yni8tlwf0bgfzc2lhbi1qaxjhlwdjltiwmtctmtatmjzfmdaxmi5sb2cumc5jdxj 3/9

http://gceasy.io/diamondgc-report.jsp?p=yxjjagl2zwqvmjaxny8xmc8yni8tlwf0bgfzc2lhbi1qaxjhlwdjltiwmtctmtatmjzfmdaxmi5sb2cumc5jdxj 4/9

http://gceasy.io/diamondgc-report.jsp?p=yxjjagl2zwqvmjaxny8xmc8yni8tlwf0bgfzc2lhbi1qaxjhlwdjltiwmtctmtatmjzfmdaxmi5sb2cumc5jdxj 5/9

http://gceasy.io/diamondgc-report.jsp?p=yxjjagl2zwqvmjaxny8xmc8yni8tlwf0bgfzc2lhbi1qaxjhlwdjltiwmtctmtatmjzfmdaxmi5sb2cumc5jdxj 6/9

G1 GC Statistics Young GC Concurrent Mark Cleanup Mixed Full GC Remark initial-mark Total http://gceasy.io/diamondgc-report.jsp?p=yxjjagl2zwqvmjaxny8xmc8yni8tlwf0bgfzc2lhbi1qaxjhlwdjltiwmtctmtatmjzfmdaxmi5sb2cumc5jdxj 7/9

Count 2529 72 72 111 1 72 72 2929 Total GC Time 3 min 5 sec 211 ms 0 1 sec 900 ms 18 sec 870 ms 290 ms 9 sec 460 ms 8 sec 710 ms 3 min 44 sec 441 ms Avg GC Time 73 ms 0 26 ms 170 ms 290 ms 131 ms 121 ms 77 ms Avg Time std dev 91 ms 0 9 ms 59 ms 0 44 ms 45 ms 90 ms Min/Max Time 0 / 2 sec 870 ms 0 / 0 0 / 70 ms 0 / 310 ms 0 / 290 ms 0 / 370 ms 0 / 280 ms 0 / 2 sec 870 ms Avg Interval Time 21 sec 357 ms 11 min 44 sec 806 ms 11 min 44 sec 808 ms 3 min 32 sec 746 ms n/a 11 min 44 sec 806 ms 11 min 44 sec 771 ms 1 min 34 sec 988 ms Object Stats (These are good metrics to include in your performance reports) Total created bytes 21.62 tb Total promoted bytes 72.69 gb Avg creation rate 419.84 mb/sec Avg promotion rate 1.38 mb/sec Memory Leak No major memory leaks. (Note: there are 8 avours of OutOfMemoryErrors (https://tier1app. les.wordpress.com/2014/12/outofmemoryerror2.pdf). With GC Logs you can diagnose only 5 avours of them(java heap space, GC overhead limit exceeded, Requested array size exceeds VM limit, Permgen space, Metaspace). So in other words, your application could be still su ering from memory leaks, but need other tools to diagnose them, not just GC Logs.) Consecutive Full GC None. Long Pause None. Safe Point Duration (To learn more about SafePoint duration, click here (https://blog.gceasy.io/2016/12/22/total-time-for-which-application-threads-were-stopped/)) Not Reported in the log. http://gceasy.io/diamondgc-report.jsp?p=yxjjagl2zwqvmjaxny8xmc8yni8tlwf0bgfzc2lhbi1qaxjhlwdjltiwmtctmtatmjzfmdaxmi5sb2cumc5jdxj 8/9

GC Causes (What events caused the GCs, how much time it consumed?) Cause Count Avg Time Max Time Total Time Time % G1 Evacuation Pause 2682 62 ms 590 ms 2 min 45 sec 982 ms 73.95% Others 216 n/a n/a 55 sec 969 ms 24.94% GCLocker Initiated GC 18 96 ms 240 ms 1 sec 720 ms 0.77% G1 Humongous Allocation 4 73 ms 90 ms 290 ms 0.13% Metadata GC Threshold 5 48 ms 80 ms 240 ms 0.11% Evacuation Failure 3 73 ms 80 ms 220 ms 0.1% System.gc() calls 1 20 ms 20 ms 20 ms 0.01% Total 2929 n/a n/a 3 min 44 sec 440 ms 100.01% Tenuring Summary Desired Survivor Size: 616.0 mb, Max Threshold: 2 Age Survival Count Average size (kb) Average Total 'To' size (kb) age 1 2710 31575.54 31575.54 age 2 2670 13752.89 45449.62 Command Line Flags -XX:GCLogFileSize=20971520 -XX:InitialHeapSize=17179869184 -XX:InitialTenuringThreshold=2 -XX:MaxHeapSize=17179869184 -XX:MaxTenuringThreshold=2 - XX:NumberOfGCLogFiles=5 -XX:-OmitStackTraceInFastThrow -XX:ParGCCardsPerStrideChunk=32768 -XX:+ParallelRefProcEnabled -XX:+PrintGC - XX:+PrintGCDateStamps -XX:+PrintGCDetails -XX:+PrintGCTimeStamps -XX:+PrintTenuringDistribution -XX:ReservedCodeCacheSize=1073741824 - http://gceasy.io/diamondgc-report.jsp?p=yxjjagl2zwqvmjaxny8xmc8yni8tlwf0bgfzc2lhbi1qaxjhlwdjltiwmtctmtatmjzfmdaxmi5sb2cumc5jdxj 9/9