application performance management vs application performance monitoring

What Is Application Performance Monitoring and Why It Is Not Application Performance Management

Matt Watson Developer Tips, Tricks & Resources, Insights for Dev Managers Leave a Comment

Over recent years, the term APM has become increasingly used by lots of vendors and tools. Some refer to it as Application Performance Monitoring and some as Application Performance Management. But what is the difference? As the founder of a company who creates these types of tools, I definitely have some opinions on Application Performance Monitoring vs Application Performance Management. It’s important to know the difference if you are planning to use an APM tool to troubleshoot app issues and need more visibility into performance bottlenecks.

What is Application Performance Monitoring?

I think you could broadly define it as anything that has to do with monitoring the performance of a website or application. For example, there are tools that do nothing but check your website every minute to see if it is online and how long it takes to load. I guess you could say this is the absolute simplest form of application performance monitoring.

However, a simple HTTP ping type check is hardly enterprise grade or advanced application monitoring. It might be able to tell you that your site is down or slow, but you would have no idea why.

Here are some other examples of basic application performance monitoring:

  • Monitoring the CPU of your servers.
  • Parsing your web server access logs to see how many requests you are getting and how long they take on average.
  • Tracking and monitoring application error rates.
  • Monitoring network traffic to identify slowdowns.
  • Tracking key metrics from app dependencies like SQL, Redis, Elasticsearch, etc.
  • Using Google Analytics to alert you about slow page speeds.

Yes, believe it or not, you could just use Google Analytics to alert you when your site is so slow.

Google Analytics Alerts

Google Analytics Alerts

The problem with most of these strategies is they don’t tell you WHY your app is down or slow. They are merely alerting mechanisms to tell you there is a problem.

Like anything in life, identifying a problem is relatively easy. Knowing about the problem is no doubt very important and these tools can help.


Free Download

That said, solving the problem is where all the hard work lies. Solving the problem takes a lot more insight, data, and tools.

How is Application Performance Management Different?

Hopefully, you can see why the term application performance monitoring can be used to mean pretty much anything. It is an umbrella term used to describe a wide range of tools that alert you about potential application problems. Unfortunately, they typically only tell you that a problem exists and don’t understand the full context of what is going on across your entire environment.

Application Management

The first big key difference is being application aware and focused. Most application performance monitoring tools measure performance in one way or another but they rarely understand the full context of the application itself.

For example, checking if https://stackify.com works or not can only tell you if the site is up or down. It has no context that the site might be running across a dozen load balanced servers.

Most APM solutions automatically discover and help you track all of your applications across all environments. Including pre-production and production environments.

There are several things that development teams need to track and manage their applications:

  • Application type (web, background service, etc)
  • Installed locations & number of server instances
  • Deployment history
  • Application availability/uptime (SLAs)
  • Web application traffic/usage
  • Performance indicators (response times, satisfaction scores, etc)
  • Error rates

These details are important to help developers understand what is going on with their applications. Thanks to the DevOps culture and affordable APM solutions, development teams finally have the tools they need to manage their applications.

Code Level Performance

Most Application Performance Management solutions also include code level performance data. The only way to identify slow SQL queries, long running web service calls and other common problems is to profile the code itself.

Most solutions, like Retrace, automatically track the performance of all common application dependencies and frameworks. Common dependencies like SQL, MongoDB, Elasticsearch, Redis, and many others are automatically recognized and tracked. This makes it really easy to quickly identify performance problems in your application stack.

Retrace Performance View by Application Dependencies

Retrace Performance View by Application Dependencies

Transaction Snapshots or Tracing

Understanding performance down to the code level also makes it possible to literally visualize what your code is doing as well. Most APM solutions can capture snapshots or traces of individual transactions. You can then review those to visualize what your code is doing. This helps developers get down to the low-level details that they need to know what their application code is doing.

Application monitoring trace

Example Transaction Trace From Retrace or Prefix

Summary

I would argue that application performance monitoring is part of application performance management. There is no real debate of application performance monitoring vs application performance management. Many of the monitoring tools and capabilities are simply features of a larger application performance management strategy.

Our product, Retrace, provides a holistic view of all the data developers need in one place. Developers can quickly access server & application metrics, application errors, logging, and even code level performance data. Developers can “Retrace” what their code is doing to quickly identify problems and opportunities for improvement.

 

About Matt Watson

Matt is the Founder & CEO of Stackify. He has been a developer/hacker for over 15 years and loves solving hard problems with code. While working in IT management he realized how much of his time was wasted trying to put out production fires without the right tools. He founded Stackify in 2012 to create an easy to use set of tools for developers.