Welcome!

Java Authors: James Carlini, Dana Gardner, Elizabeth White, Gary Kaiser, Noel Wurst

Related Topics: Java

Java: Blog Feed Post

Are You Riding the Java Performance Rollercoaster?

It would be wonderful to tell when a stock was going to go up before it did

We all wish we could predict the future.  It would be wonderful to tell when a stock was going to go up before it did.  Unfortunately, most of us are not psychic.  In order to be successful in the stock market, you need to be lucky or spend a lot of time analyzing the behavior of the stocks.  Wouldn’t it be valuable if you were able to predict the up and down performance behavior of your Java applications and make adjustments to avoid unfavorable impacts on your company? 

We find that one of the biggest impacts of not being able to predict application behavior, is that companies miss their SLAs.  This can result in fines or other penalties.  Chances are slim that with the hundreds of Java applications that our companies have running at the same time that the IT group will be able to monitor each of these applications and predict when there will be a problem.  Users are generally the ones that detect the performance problems.  Response time problems and failures affect everyone and are not only costly, but time consuming as well.  No corporation can afford to have erratic performance or down time, especially not on a regular basis.

We have seen that when you have multiple JVMs running, the need for performance trend analysis increases dramatically.

Of the different methods, I find that these three simple methods that can be used to detect performance trends in applications are:

  • Bollinger Bands™ - The Bollinger Bands method determines a high and low band using the given set of samples.
  • % Change – The % Change looks at the trend from sample to sample.
  • Velocity – Velocity uses the number of measured units of change or a velocity of change per unit of time.

These 3 indicators can be used to determine short term and long term performance trends for an application.

Just like in the stock market, a good application strategy is the best offense.  Measuring the trends of your application performance behavior can make you a winner.

To find out more about how to measure the highs and lows, ups and downs, what your applications have been up to and learn how to smooth out the ride, Check out our video “3 Ways to Detect Java Application Performance Trends” or visit our website and review the content on application performance monitoring.

More Stories By Charles Rich

Charles Rich is Vice President of Product Management at Nastel Technologies, a provider of middleware-centric application performance monitoring for mission-critical applications from the datacenter to the cloud. He is a software product management professional who brings over 27 years of technical hands-on experience working with large-scale customers to meet their application and systems management requirements.

Comments (0)

Share your thoughts on this story.

Add your comment
You must be signed in to add a comment. Sign-in | Register

In accordance with our Comment Policy, we encourage comments that are on topic, relevant and to-the-point. We will remove comments that include profanity, personal attacks, racial slurs, threats of violence, or other inappropriate material that violates our Terms and Conditions, and will block users who make repeated violations. We ask all readers to expect diversity of opinion and to treat one another with dignity and respect.