Quickly Isolate Workload Bottlenecks

Hi, folks. In this video, we're going to show how you can quickly isolate workload bottlenecks using foglight for databases. First of all, from the Home screen, select the instance in question. Then select SQL PI. First of all, you need to set your date and time range. You can do so by setting a date and time in the from and to fields, or selecting one of the preset time ranges. Then we're going to select Activity Highlights. This will show us the worst offending components for the selected time period, such as SQL statements, [? T-SQL ?] batches, or [? PL/SQL ?] code, databases, programs, users, and objects. You'll also see a resource breakdown and top wait event for each component. Each component is also hyperlinked so you can click on it to get more information. Going back to the overview, we can see for the selected time period the instance spent 44% of time on the other wait event. So what I want to do is isolate this and see the workload that contributed towards it. So I'm going to select the other dimension. This will filter the screen to show us information just for the selected wait event. Now using the performance tree, I'm going to try and isolate the workload. First of all, I'm going to see the list of databases. And we can quite easily see that the vast majority of wait time can be attributed to the paused database. So I'm going to drill down and select a paused database. Now I'm going to see what users I can attribute the wait time to. And again, you can see the vast majority of wait time can be attributed to a single user. Now I'm going to drilldown in this user and see what programs [? we're ?] running the workload. You can see it all came from a single program. So now I can see a list of SQL statements run through this program by this user on this database that generated other wait events for the selected time period. So that's how we can quickly isolate workflow bottlenecks using foglight for databases. For more information on foglight for databases, go to quest.com forward slash foglight for databases. Thanks for tuning in.