APM tool framework makes sure you cover the bases
If you love Wireshark, hunting down performance problems at the packet level gives you an endorphin buzz. You crave hard problems and the raw data to solve them in your head. But you can’t always parachute in with a Sniffer when a problem pops up. For that you need a tool that loves you back by incorporating some of your gray matter. This doesn’t mean forsaking hands-on tools and doing yourself out of a job–it means deploying the right set of tools to make your job easier and make the most of your smarts. But how do you know which tools will love you in a way that leads to a satisfying long-term relationship?
We’ve thought a lot about this problem, and determined that the key to choosing the right tool set is to place your APM tool choices into a “big picture” context (or framework) that helps you define what you need from your tools, and map those needs to the current crop of tools.
Here’s the underlying concept of the NetForecast APM Tool Framework. There are three application performance management functions:
1) Gathering performance data,
2) Interpreting the data, and
3) Using the results to support the organization.
Gathering the Data:
There are two key aspects to gathering application performance data–technique and location. Technique determines how to measure performance, and location determines where to measure it.
Interpreting the Data:
Data interpretation involves first determining what to measure (i.e., what data to gather), and then to analyze the data to make sense of it.
Using the Data:
When all is said and done, the results of the analysis must support individual processes–and ultimately the business as a whole. For information to be useful it must be: based on analysis of performance by time of day, location, and business function; trend analysis over long periods; integration and correlation of data from multiple sources (e.g. other tools and outside data feeds); and it must be stored in a long-term data repository for data mining.
The Wiresharks of the world are good on the front end of this functional flow, but they require lots of human gray matter to make sense of the data, put it to use, and keep a record of it. Although there is no substitute for a smart geek, there is much to be gained by incorporating geek-smarts into automated tools that analyze data and apply it to improve application performance.
The back end of the functional flow (analysis, process, business reports) is where the tool makes your life easier and shows value to your management on a daily basis. You are not going to study packet traces for performance issues every day. But a tool that automatically shows performance information every day changes your job. It shows that you are confident in your abilities and that you place your business’s needs above your personal pride.
Let your tool love you back. It will free you to do more interesting things while the tool generates a daily report of how well you are managing application performance. But finding a comprehensive APM tool that meets both your needs to investigate and your management’s need to understand performance is not simple.
Stay tuned as we will be providing information on all of these aspects of the NetForecast APM Tool Framework in subsequent blogs. Our goal is to help you find the right APM tool lineup so you assemble the right mix of tools for your unique needs.
