SLA Monitoring: Using Internal SLA's to Improve Application Performance

NOTE: ExtraHop no longer offers the NetworkTimeout.com service. These links now lead to our online interactive demo.

Most of us in business are familiar with the classic adage: "if you can't measure it, you can't manage it." As IT becomes more deeply aligned with the business it supports, the ability to understand and measure key performance metrics around IT services has grown significantly in importance. Last week, Michael Biddick authored an article in InformationWeek around the issue of keep[ing] users happy with service-level agreements and application performance management.

His premise is simple enough: companies have relied on service level agreements (SLAs) for years now to ensure that acceptable service levels are clearly laid out and agreed to by both parties. This not only provides accountability but also articulates consequences if agreed upon service levels are not adequately met. So why not apply that same principal to internal departments that are responsible for delivering services? This is precisely what many forward-thinking enterprises are now doing to better understand not only their own cost structures, but also how different departments are leveraging internally delivered services.

As Michael points out, "lots of IT organizations use an informal metric for measuring how satisfied employees are with application performance: If they aren't calling the help desk, things must be OK." As is often the case, the vast majority of employees aren't going to make much of a fuss over sluggish application performance and consequently, IT often doesn't know there's a problem—a problem which is likely becoming worse. This is why it's becoming a priority for IT to understand where applications are failing as early as possible to ensure that network hiccups don't cascade into application downtime.

Using the ExtraHop Application Delivery Assurance system, you can take a number of steps to ensure that SLAs are achieved, including:

• Alert on problems before SLA violations occur. The ExtraHop system sends notifications when issues like network slowdowns, applications errors, and slow DB processing times are detected.
• Fix intermittent issues. The ExtraHop system passively analyzes all transactions on the wire in real time to isolate intermittent failures and slowdowns, improving service assurance.
• Proactively detect anomalies. The ExtraHop system learns normal behavior based on historical context and provides trend-based alerts to warn of potential problems.
• Increase service transparency. If you can't measure it, you can't manage it. The ExtraHop system can baseline and analyze services passively with no overhead at all. Try it for yourself at NetworkTimeout.com.