Good morning. My name is Steven Stout. I am a pre-sales consultant working for IDERA. Today we are going to be short video just to look at the feature of setting up thresholds for alerts and setting up the baselines related to them. So to start with how we can set up the actual baselines themselves. If you come in here to the properties you will see all the properties related to a specific instance. Today we are just going to look at the baseline. And this will take us indirectly to the properties page. Looking at the specific area of interest. Now the variety of ways you can actually set baselines within IDERA. You can go to automatically calculate dynamic baselines based on the last seven days’ worth of collected data. And this is the normal way that most of our clients do it. You can if you wish by the way use a specific date range if you know you have a representative set of data that will be appropriate. And then within the baselines, you have the default baseline. This is the one that is used in most situations. Whereas you can see here I have got it set Monday to Friday during the working day. But then I can also create additional baselines looking at different time ranges or different dates on different days. And that can be very useful if you have got perhaps OLTP [online transaction processing] action happening during the middle of the day. And then in the evening, you have got decision support reports and things which are creating very different loads on your system. And you actually want to use or collect the information and give you [… ] baselines based on the different characteristics of those two types of workload. When you want to create a new baseline you can. Let us just click on here. Give the baseline a name and a date rate time day and time range. And add it into the system. And here is an example of where I have actually created evenings during the weekday working from 5:00 to just about 10 past 11:00 in the evening. Having created those baselines then we can actually see how you use them. So if I come to look at the instance itself. And though that the figure alerts. And this is actually how we are setting up what we want to alert on for this instance. And setting the thresholds for that. So as you can see as before there is a wide range of things that IDERA can actually alert off or load on. Rather one of those I am just going to use today is in the resources area. And we are looking here at the OS [operating system] memory usage. So using the traditional way that IDERA would look at this sort of information this is actually showing me that I have got two levels of threshold set. I have currently got it set to an unrealistic threshold for my warning of 20%. And I can easily just kind of drag and drop that around to set the threshold as I always see a […]. Or I could type the value in here. But you may notice that beneath the line we have this green hatched area. Here this is actually telling me what the baseline is for this matrix. And it is summer between 70 and 85 and that is the baseline the default baseline as calculated when we saw […] on the settings a few moments ago. The other very useful thing down here is a baseline threshold enabled as a percentage. What this is then saying is we want the threshold to trip based on a certain percentage of what the baseline value is at the point in time when we are actually comparing the metric. So if I collect this metric I would say in the middle of the day I am going to compare it against whatever the baseline value is for the middle of day range. So that is my default in this context. Whereas if this metric is being collected let us say at 7 o’clock at night in the middle of our report run and then maybe I want to actually prepare this against what is normal for the baseline value at that time of day. So what this like to do is then trigger of mere warning if I am at a hundred percent of whatever the baseline value for that point in time. And similar I have got done here that it was 120 percent of what is normal. I have had a period of time then I want to trigger this alert.
Topics : Database Monitoring,Database Performance,
Products : SQL Diagnostic Manager for SQL Server,