https://github.com/ionutbalosin/faster-jvm-start-up-techniques
App/Dynamic CDS, Shared Class Cache SCC, Ahead-of-Time AOT, OpenJDK HotSpot, Eclipse OpenJ9, Graal VM native-image
https://github.com/ionutbalosin/faster-jvm-start-up-techniques
appcds cds dynamiccds graalvm-native-image hotspot-jvm jvm openj9
Last synced: 28 days ago
JSON representation
App/Dynamic CDS, Shared Class Cache SCC, Ahead-of-Time AOT, OpenJDK HotSpot, Eclipse OpenJ9, Graal VM native-image
- Host: GitHub
- URL: https://github.com/ionutbalosin/faster-jvm-start-up-techniques
- Owner: ionutbalosin
- Created: 2022-03-24T15:14:07.000Z (about 3 years ago)
- Default Branch: main
- Last Pushed: 2022-10-06T10:16:20.000Z (over 2 years ago)
- Last Synced: 2025-04-01T18:17:39.900Z (about 2 months ago)
- Topics: appcds, cds, dynamiccds, graalvm-native-image, hotspot-jvm, jvm, openj9
- Language: Shell
- Homepage:
- Size: 11.7 KB
- Stars: 52
- Watchers: 2
- Forks: 7
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Faster JVM Start-up Techniques
Techniques about how to improve the JVM start-up time for any application running on the JVM.
Please use this tutorial in conjunction with my article [Application / Dynamic Class Data Sharing In HotSpot JVM](https://ionutbalosin.com/2022/04/application-dynamic-class-data-sharing-in-hotspot-jvm/) and my presentation slides [Techniques for a faster JVM start-up](https://ionutbalosin.com/talks).
>Visit my [web site](https://ionutbalosin.com/) if you are interested in further performance or software architecture-related articles.
>Check my [training catalog](https://ionutbalosin.com/training) if you are interested in specialized training for your company.
## Content
#### [App/Dynamic Class Data Sharing (CDS) in HotSpot JVM](app-dynamic-cds-hotspot/README.md)
#### [Shared Classes Cache (SCC) and Dynamic Ahead-of-Time (AOT) in Eclipse OpenJ9 JVM](scc-dynamic-aot-openj9/README.md)
#### [Ahead-of-Time (AOT) with GraalVM native-image](aot-graalvm-native-image/README.md)
#### [App/Dynamic Class Data Sharing (CDS) with Docker - WIP](docker/README.md)
## Projects in scope
### spring-petclinic project
```
git clone [email protected]:spring-projects/spring-petclinic.git
```
Additional files are needed for the multi-stage Dockerfile but also to measure the application start-up time
```
cp ./additional-scripts/* ./spring-petclinic
```
### spring-native project
```
git clone [email protected]:spring-projects-experimental/spring-native.git
```
Additional file is needed to measure the application start-up time
```
cp ./additional-scripts/time-to-first-response.sh ./spring-native/samples/petclinic-jpa
```
## Disable Address Space Layout Randomization (ASLR)
For comparable results (especially in the case of App/Dynamic CDS), please make sure [ASLR](https://en.wikipedia.org/wiki/Address_space_layout_randomization) is disabled.
For further details check [JEP 310: Application Class-Data Sharing](https://openjdk.java.net/jeps/310)
```
sudo sh -c "echo 0 > /proc/sys/kernel/randomize_va_space"
```
## Check the process memory
### Resident Set Size (RSS) / Proportional Set Size (PSS)
When multiple JVM instances are running on the same host and the archives are shared, the overall memory is implicitly reduced. To measure it I recommend looking at the [RSS](https://en.wikipedia.org/wiki/Resident_set_size) and [PSS](https://en.wikipedia.org/wiki/Proportional_set_size) of each process by using the [pmap](https://www.labcorner.de/cheat-sheet-understanding-the-pmap1-output/) command (known to provide the most accurate information) .
Assuming there are 2 running processes and PIDs were incrementally assigned
```
pmap -X `pgrep -f petclinic | sed -n -e '1p'` | sed -n -e '2p;$p'
```
```
pmap -X `pgrep -f petclinic | sed -n -e '2p'` | sed -n -e '2p;$p'
```
### Heap and Metaspace statistics (in HotSpot JVM)
```
java -Xlog:cds -Xlog:gc+heap+exit -jar -Dserver.port=8080 target/*.jar
```
```
java -Xlog:cds -Xlog:gc+heap+exit -jar -XX:SharedArchiveFile=dynamic-cds.jsa -Dserver.port=8080 target/*.jar
```
Once the application properly starts press **CTRL+C**.
**Note:** Check the used/committed memory