A common refrain that coaches and experts repeat is the importance of all three phases of the game and how all the units of the team need to work together to succeed. Watching games, it's easy to see that in action: Teams with good offenses but lousy special teams and defense can put up a lot of points but often don't make it to the postseason or don't last long if they do.
But this is a lesson that, for a while, hasn't taken hold in the application and network performance infrastructures of many businesses. In fact, it has been all too common for a company to have an application team that worries just about testing and has little concern for network issues. Or network admins who are looking at only network capacity and not at how performance affects the user experience.
Clearly, these organizations need a good head coach who can whip them into shape and ensure that all groups are working together to achieve the highest levels of application performance. That's because many of these same team-oriented principles apply just as much, if not more, in application performance management as they do in football.
When an application suffers from a performance or usability issue, the cause could be in the code, or due to a server or network issue, an interface design flaw, or some combination of the above. The code flaw could be found by testing tools used by the development team, but they would be ignorant of the server issues that would be readily apparent to the data center team.
The first step to overcoming this problem is to get the different teams to stop their siloed approaches and start working together. But this solves only part of the problem.
The other issue is that the different teams don't really understand or know how to get the most value out of the performance management and monitoring tools in use by each group. Testing and debugging tools can be as hard for network admins and business managers to understand as a quarterback's playbook would be to a defensive tackle.
And while business managers might understand that red lights are bad in a network monitoring console, they won't really know how to figure out what the underlying issue causing the red lights is. That's why the next step is to improve translation layers and integration among the different performance management tools in use by each group. By enhancing the flow of information, this can make it possible for the different teams to work together to understand application performance and detect and remove potential issues.
And these steps can pay off in improved application performance and increased monitoring and management capabilities. In my recent report, Overcoming Application Performance Walls, survey data showed that businesses that have a unified approach to performance management are 75% more likely to have implemented capabilities such as identifying sources of delay in application performance. And this approach should become more common in businesses, as 88% of organizations centralize management of application performance, or plan to in the next 12 months.
In football, when the offense fumbles the ball, it isn't just their problem. The defense has to go out and limit the damage of the turnover, and special teams needs to ensure good field position to get the offense back on track.
Similarly, when an application fails, it isn't just the application development team's problem. When all of the groups within an IT infrastructure are able to work together to manage performance, everyone wins.