Insight is your starting point when going after a performance issue. Built upon a patented correlation engine, Insight looks at all of the tiers in your application stack and ranks them according to the amount of resources that each is consuming. Insight tells you where to look on the very first screen! A very nice tool for the Help Desk to begin problem resolution.

Once you know where to look, the appropriate expert uses Indepth that provides fine-grain, highly-detailed drill-down capabilities into the problematic tier to determine the root cause, Usually within just 2 or 3 mouse clicks Indepth brings you to the statement that is causing the problem. It also provides expert advice on how to fix the problem.

So that you don't have to sit and watch your console all day long looking for performance problems, Inform automatically notifies the appropriate application stakeholder that there is a critical problem, or even a near-critical problem which puts you on the way to being proactive in managing performance. Inform also provides trending reports.

Here is an example of
Insight's Actionable Performance Data

Even with just a cursory glance at this chart, immediately shows that Tuxedo is consuming the most resources, with Java not far behind. The lower portion of the screen shows all of the root activities (e.g. URL's, methods, SQL, etc) running at the time and how they behaved.

In this case, if we were having a performance problems during the time frame shown, the DBA's would the LAST group we would go to. Also looks like Web is "off the hook" too.

Here is an example of
Indepth for SQL Server

The upper half of this screen shot shows the overall activity at the database level.

In the lower half, we are seeing the individual statements and how they were performing. Based on the colors, we can see a significant amount of I/O Wait, and Latch Wait.

Here is an example of the
Inform Alerts Dashboard

An important feature of i3 is that it automatically calculates a baseline of normal application activity. This is very powerful when configuring the thresholds for the multi-level Inform Alerts.

For example, a normal query usually runs in .25 seconds. The SLA is 1.50 seconds. When that query starts to run at .85 - 1.00 seconds, this is a good point to issue a "Near-Critical" Alert to indicate that something bad is about to happen. You can then fix it before the user finds it.