Slides: JVM Performance – Past, Present and Predicted

Over the last 3 weeks I have given four talks on JVM performance monitoring and optimization in terms of past, present and predicted techniques, technologies and trends. The talks have taken place in Kuala Lumpur (Malaysia Java User Group), Sydney (Clojure User Group) and Melbourne (Scala User Group, Java/JVM User Group). Last night I gave the final talk in front of a +50 Java developers at the Melbourne Java/JVM User Group monthly meetup at PeopleBank coordinated largely by Kon Soulianidis on this particular occasion. It was a great night though I was pushed to get through so much material in constructing the argument for a future of software that is self adaptive and self regulated.

Here are the slides I used on the final night.

Below are links to articles which provide further information on the concepts introduced and topics covered in the slides.

The Complexity and Challenges of IT Management within the Cloud
Feedback & Control signals its arrival in the Cloud & Enterprise
Visions of Cloud Computing – PaaS Everywhere

Dynamic Metering “Overhead” Budgeting
Controlling Runaway Threads in the JVM using Resource Metering Quotas
From Anomaly Detection to Root Cause Analysis via Self Observation

Using System Dynamics for Effective Concurrency & Consumption Control of Code
Optimal Application Performance and Capacity Management via QoS for Apps
QoS for JVM Applications – An Introduction

Prediction is the Future in Application Performance Management & Software Optimization

Where Big Data meets Big Activity in the JVM

It’s the Application not the Process that’s important
Time, Space & Data in Application Performance Monitoring

If you are interested in having a similar talk with your local Java/Scala/JVM User Group you can contact me using the email address listed on the first page of the slides.