In the first part of this article series, I showed you how to use the System Center Capacity Planner to create a model that can be used to plan a Microsoft Operations Manager deployment. Although we created the model, the model was rather generic. Unless your network is really basic, there's a good chance that the model only loosely resembles your real world network. In this article, I will conclude the series by showing you how to use the Model Editor and the Hardware Editor to fine tune the model that you created in part one. I will then go on to show you how to run a deployment simulation using the model that you have created.
The Model Editor
The model editor's purpose is twofold. Initially, you will want to use the model editor to make your model more closely resemble your real-world network so that you can see the impact that a proposed MOM deployment will have on your network. When you eventually run the simulation, it is possible that you may not like the results that you are given. If this happens, then you can use the Model Editor to tweak the model. By doing so, you can test the impact of alternate hardware or an alternate deployment topology. You can keep making changes to the model and running simulations against those changes until you get the results that you want.
Begin the process by opening the System Center Capacity Planner. When the Capacity Planner opens take a look at the options on the Welcome screen. The second option from the bottom allows you to edit an existing model with the Model Editor. Select this option and load the model that you created in Part 1. You'll now see your model opened within the Model Editor.
AThe middle pane displays the network's global topology. At first glance, the screen does not appear to give you much information. However, if you hover your mouse pointer above any of the offices, a brief summary of that office will be displayed. The summary includes such information as the name of the site, the number of servers in the site, the number of clients in the site, and the site's usage profile.
The Global Topology screen is primarily used to edit general aspects of the network. For example, if you look in the lower right hand corner of the screen, you'll see options for tasks such as adding a new site, editing a site's usage profile, or adding a new WAN.
Since you have just created the model, the model's global topology should be a fairly accurate representation of your real-world network. The one thing at the global level that probably isn't accurate though is the speed of your WAN links. As you may recall, when you initially created the model five Capacity Planner asks for your WAN connection speed, and of the saturation limit for the link. The problem is that the Model Creation Wizard assumes that each of your WAN links is running at the same speed and has the same saturation limit. You will therefore want to review the speed of each link and make any necessary changes.
Reviewing and modifying link speeds is easier than you might expect. All you have to do is double-click on the link that you want to review. When you do, you will see a screen similar to the one that's shown in Figure A.
|The Model Editor allows you to customize each of your model's WAN links.|
As you can see in the figure, you can set the connection's uplink and downlink bandwidth. As I explained in the first part of this article series, you don't want for MOM to consume all of the connection's bandwidth. Therefore, you should enter the available bandwidth as a percentage of the total bandwidth. The value that you choose reflects the maximum amount of bandwidth that you want MOM to consume. You also have the option of adjusting the connection latency if necessary. After you have made any necessary corrections to the link, click OK and then repeat the process for all the remaining links.
Now that we have corrected the model's WAN link speeds, let's move on to editing the one of the individual sites. The technique that you will use for editing a site is the same regardless of which site you're working with. For the purposes of this article though, double-click on the Central Office site. When you do, you'll see that the Model Editor switches from the Global Topology screen to the Site Topology screen, as shown in Figure B.
|The Site Topology screen allows you to make modifications to an individual site within your model.|
As with the Global Topology screen, the first thing that you'll probably want to check is the speed of the links between the various components. To do so, just double-click on a link like you did earlier.
The next thing that I recommend doing is editing the server roles. During the Model Creation Wizard, you were asked which roles needed to be performed. The wizard incorporated the specified roles into the model, but did not necessarily associate those roles with the correct servers. Therefore, I recommend taking a look at which roles were assigned to which server, and making any necessary corrections.
You can see the roles that are assigned to a server by hovering your mouse over a server's icon. When you do, you'll see a summary of the server's configuration. Most of this information is hardware related (I will talk about hardware later), but the bottom part of the summary lists the server's roles. If the roles assigned to a server are incorrect or incomplete then click on the server and then click the Edit Server Role option in the Current View Actions section. You will now see a screen similar to the one that's shown in Figure C.
|You can customize the roles that are hosted on each server.|
As you can see in the figure, you can enable or disable the various roles on a server simply by selecting the appropriate check boxes. You might also notice in the figure that when you select a role the application components associated with that role are displayed just to the right of the role. If you select an application component, you can use the options in the Disks section to choose which hard disks or disk volume hosts that particular application component.
In a moment I will show you how to use the Hardware Editor to customize a server's hardware. Although the Hardware Editor allows you to control a server's disk configuration, it does not allow you to choose which disk each application component is using. You may therefore have to return to the screen later on.
The Hardware Editor
When you initially ran the Model Creation Wizard, it asked you to pick up to three different server configurations. When the model was created, the server configuration or configurations that you have chosen are distributed throughout the model in a seemingly random way. That being the case, it is highly unlikely that any of the servers in the model accurately represent the hardware configurations of their route world counterparts. This is where the Hardware Editor comes into play. The Hardware Editor allows you to customize each server in your model so that the modeled hardware matches your real life hardware.
The process of configuring hardware within the model is a little bit unusual. You must initially designed a configuration that matches your real life hardware. After you've created such a configuration, you must go back into the Model Editor and assign the configuration to one of the servers in the model.
To create a hardware configuration, click the Hardware Editor button found in the lower left corner of the screen. When you do, you'll see a screen similar to the one shown in Figure D. If you look at the Computer Configurations pane in the figure, you'll see that the top half of the pane shows a list of all existing configurations.
|The Computer Configurations pane lists all existing configurations.|
Assuming that your server hardware isn't listed among the existing configurations, you'll need to define it. To do so, click the New Computer button. When you do, you will see a screen that looks something like the one that's shown in Figure E.
|The Hardware Editor allows you to define a new computer configuration.|
If you look at the bottom of the screen, you'll see a list of various processor configurations. Select the configuration that most closely matches one of your real life servers, and click the Add Device button. When you do, the chosen processor will be added to the Devices in the Computer Configuration section of the screen. Now select either the Single Disk or the Disk Array option from the View drop down list. The Hardware Editor will now display a list of various hard disk or disk array configurations. Choose the disk configuration that most closely matches your real-world server and click the Add Device button. Repeat the process if necessary to define additional hard disks.
As you look through the list of CPUs and hard disks you may find that options matching your real life hardware are simply unavailable. If that's the case, you do have the option of defining custom devices. If you look at the above figure, you will notice that the Computer Configurations option is selected in the upper left-hand corner of the screen. If you choose the Device Configurations option instead, you will be taken to the screen shown in Figure F.
|The Device Configurations page allows you to configure custom CPU and disk configurations.|
As you can see in the figure, defining a custom device is simple. Just select the device type from the View drop down list, and click the New Device button. You are now given the chance to enter all of the specifications for the device that you're defining.
Implementing Custom Servers
Now that I have shown you how to create custom hardware and custom devices, it's time to implement your custom servers into your model. Begin by clicking the Model Editor button found in the lower left corner of the screen. When you do, you'll see a message stating that you have not saved to the changes that you made in the Hardware Editor. Click Yes to save your changes and you will be taken back to the Global Topology screen of the Model Editor.
Now, double-click on the site containing the server that you want to modify. When the Site Topology diagram appears, double-click on the server that you want to modify. When you do, you'll see a screen similar to the one that's shown in Figure G.
|The Model Editor allows you to customize your server's hardware.|
If you look at the bottom portion of the figure, you'll see a couple of fields that you can fill in. The first field is the Server Name field. I recommend entering the NetBIOS name of the server that you are simulating. You don't actually have to enter a server name, but entering a server's real life name helps to make the diagram easier to follow, and helps avoid confusion.
After you enter the server's name, select the appropriate hardware configuration from the Apply New Configuration drop down list and click OK. The model will now be updated with the server's new name and configuration.
Running a Simulation
After you have finished customizing your model, it's time to run a simulation against the proposed MOM deployment. Running a simulation is really simple to do. All you have to do is click the Run Simulation button the top of the screen. When you do, the simulation will begin immediately. The length of time that the simulation takes to complete varies depending on the speed of your computer and the complexity of your model. Most of the times that I have used the System Center Capacity Planner though, the simulation completes in under 10 seconds.
When the simulation does complete, you'll be presented with a screen similar to the one that's shown in Figure H. As you can see in the figure, the System Center Capacity Planner provides you with a wealth of statistics regarding CPU and storage utilization and transaction times. If you look at the Simulation Results column on the left, you will see that there are a variety of reports that you can look at.
|The Results Summary is only one of 10 available reports.|
Hopefully all of the data produced by the simulation is favorable. If not though, just click the Model Editor button in the lower left-hand corner of the screen and you will be given the opportunity to tweak the model. With enough tweaking you should eventually be able to get the results that you want.
Worth the extra work
As you can see, the System Center Capacity Planner is a very powerful tool for testing various MOM deployment scenarios. Although there is sometimes a considerable amount of work involved in creating an accurate model, the System Center Capacity Planner can prevent you from making an extremely expensive mistake when it comes to purchasing hardware and software for your MOM deployment.