 Hello everyone, my name is Jim Waring, Solution Engineer in Performance and Analytics, and I'd like to take this opportunity to step through the new DataMart Explorer template that's being released as part of the version 11 capacity optimization solution. The new DataMart Explorer template provides us with a quick easy way in terms of visualizing the contents of some DataMarts and being able to combine together both summary information as well as time series data in a single view. The value to the customer for this is really faster adoption of the solution and publication of information. It simplifies the creation of these views across any of the data sources that are within the DataMarts, and it gives us an easier customization and look and feel for the customers as well. Before this capability, this template was created, we could only look at either system detailed information or summary data in terms of how we would build views, and they would be in different tabs or different pages within the view. What we get with the new template is a combination of both of those, dramatically simplifying how people can navigate their way through the data. With that, I'll cut straight to the chase in terms of taking you through using this template. The first thing I wanted to do was to step on the DataMarts that I'm going to use within the template. What we've got is some DataMark views. The first one is a summary view, and if I click on this and do the preview, you can see that it's given us a summary table in terms of the individual Hadoop clusters and the various distributions for those clusters. If we edit this, we can see that this particular view is being exposed to the true site console and that we are categorizing it as a summary DataMark. In comparison, if I look at the CPU DataMark, this is a time series, and I can see I've got lots of different integrals and values across time. Again, if I look at the attributes of that, I can see that it's exposed to the true site console, but it's a time series set of metrics associated with some entities. If I go into the true site console, I can now add a new view. I'll give it a name of Hadoop DataMark. I am going to save it in my own little sandpit. Now, here are all the available templates, and the one that I'm going to step through now is the DataMark one and what I'll now see is some blank areas that need to be configured. The first one I'm going to do is the Summary one, and if I click on the cog, I can then select the DataMark. If I hover over the DataMark, I then get the full description, so I can see that this is just the cluster one, where this is the CPU, and the last one is the memory, so I know I'm selecting the right DataMark. I'll call it Summary, and then I can start selecting the values to appear in that table. So I want the name, and then I'm going to select Average CPU Usage, and then I'm going to look at memory used, and then I'll look at the file system used. But what we can also do is look at the attributes for that particular piece, and rather than just publishing the number, I want it to appear as a bar, and the size of the bar is going to be determined by the amount of real memory, and rather than make it appear in bytes, I'm going to do it in gigabytes, and I want the whole number. I don't want the mantissa or just the whole number, and then I'm done on that one, and I'll do the same on here for the file system as well. So make that a bar, then the maximum value is the total amount, and this time I'll show it as terabytes, done. Apply that, and now if I just scroll down, I can see that it's filled out those bars for me. Because I've now defined that, I can also up here look at the filters that are going to appear on this page. So I want to look at filtering based on distribution. I've only got two distributions, i.e. Cloudera or Hortonworks, so it makes sense to click on the quick filter. If there was more than a six or seven, then if I don't click on this, I would then have a dropdown to make the appropriate selection. Then I can go through and say, let's have our CPU. Again, data mark that I'm going to select is CPU. I want the metric, and then in here the number of cores, apply, and then I'll do the same, but this time for memory, real memory, and apply. And now I can save this, and what we can see, let me just get a bit more real estate on here. I've now got my two definitions of distribution, and I can filter on these. So now I just see the aggregate for the Cloudera clusters, or I can deselect that and select the Hortonworks side of things. Something else in terms of this, because I'm going against the data mark, there's not much point in having the domain or the tag filter here. So if I edit the view, I can just disable those to make it a bit cleaner and save that. And then another quite interesting thing for a good capability is if I edit the page, I can go back to my table and then rather than just displaying the name, I can actually say that this is gonna be a link. And now I can select somewhere to actually do an additional drill down. So I know that I've got an Out-of-the-Box Hadoop view and I can just select my cluster detail as to where I want to want to land to. So now if I do that, apply it, save it. Now when I look here, I can see that I've got a link, a hyperlink here. So I can just click on that and it will take me straight to that additional page and give me the details for cluster B.