Identifying Capacity and Performance Trends

I’m going to take you through a couple examples of how you can use the trending facilities [in IntelliMagic Vision]. We have trending facilities for both capacity and performance.

Capacity Trends

Let’s take a look at what we can do in the capacity. And what I’m going to show you here is how the free space for a set of controllers by Tier actually changes over time.

So, we’re going to start out by looking at Free Tier over time, and we’re looking at some different tiers here. We got 450gb drives, 900gb, 300gb, and then some mixed tiers.

And, I actually want to drill down to a particular tier, so I’m going to use the isolate button, and then I’m going to drill down – I’m going to see the trend just for this tier in a line chart. It’s a little bit easier to see when we show a trend line, so I’m going to add that trend line.

You can see that the trend is actually decreasing. It starts off at about 79% full and by the end it’s at 76%, so this is actually a case where the provisioning – the actual percentage of the pool that’s in use – has decreased over time. So that’s actually a good trend.

Performance Trends

There’s other kinds of trends though, that are important to be able to see, and those are the performance trends, so I want to hop over here to the trending yearly and look at the throughput over time and see how we’re doing in relationship to throughput.

So, I’m going to isolate, again, on the biggest area – the biggest piece of the pie here – which is svc001, and that’s going to show me the trend over time.

It’s easier to see the trend with the trend line, so we’re going to add that to the chart, and you can see there’s really not much change here, maybe a slight decrease.

How about the response time for the same controller? I’d like to know what it is, so I’m just going to go over here and look at the response time, isolate on that response time, and I can see how my response time changed over time.

Really, it didn’t change much either – that makes sense since we didn’t see an increase in the workload.

My conclusion is that I can really use any of the metrics here to trend over very long terms – years even – for capacity and performance, even at an application level if I want from a performance perspective to see how workloads are changing over time, and see if there’s any concerning trends.

All of this data I can export to excel for further analysis and predictive forecasting, and using those things you can avoid negative trends.