At MMS last week I presented a session looking at various reporting scenarios based on the feedback we have received since we launched R2 last year.
The purpose of the session was to try and address the common problems and show the options available to Operations Manager users with regards to report creation. Some of the feedback we were looking to address is highlighted below:
“I need to report on how often my servers were in maintenance mode”
“Why does Visual Studio seem the only option for creating the reports I need?”
“Can we use operations manager for capacity planning?”
“Dashboard reports that show a lot of general info…..”
“How do I make reports pretty….”
We broke the session down into four main demonstrations that included samples that customers could take away and use. This blog includes those samples and instructions for using them.
Note: For those who attended the session at MMS, I stated the samples were attached to the deck but unfortunately after the deck was submitted the linked samples were removed.
• “Customer X needs to produce a report that shows the processor, memory, logical disk performance for his 100 servers and needs to see the average performance over multiple time ranges”
In this demo we showed how the Service Level Tracking feature combined with the Service Level Dashboard add-on could be used to create a self-service reporting portal.
The attached file PerformanceReporting.xml contains the sample service level rules required to create this. It also contains some additional collections rules for Disk Space targetted at the OS to make the viewing experience the same.
To use simply import the MP. This MP is targeted at Windows Server 2008 servers so to use you may need to adjust based on the servers you monitor. You may also want to adjust the service level objective goals to make them appropriate to your environment. Once imported you can go to the Service Level Dashboard and add the three service levels – Disk, Memory and Processor. This will create a portal as shown below:
• “Customer X needs to produce a report showing detailed information about how long servers spent in maintenance mode and what happened during that time”
You can download this and take the query MaintenanceModeHistory.txt in the attached file to create something like the below:
• “Customer Y is looking for physical servers in his environment that may be suitable candidates for Virtualization, he needs a report that can show overtime the most suitable servers”
Again using the Service Level Tracking features we showed how you can create business rules(service level objectives) to drive decisions such as Virtualization Candidates and view this using the Service Level Tracking Summary Report. The VirtualizationCandidates.xml in the attached file contains the configuration to do this. Simply import the MP, adjust your service level objectives from the service level tracking node under authoring in the console and run the Service Level Tracking Summary Report.
The Service Levels in this report are scoped to particular groups based on server hardware configuration. You should create your own groups based on your own hardware and link to Service Levels. The groups in the attached MP are Medium SKU Servers and Low SKU Servers and are targeted to Windows Server 2008. Once the MP is imported open the two service levels and on the scope settings page change to your own groups or reselect the two groups above:
Change to correct group:
“Customer X need to provide to his IT management proposed requirements for new hardware, he would like to produce a report that shows what servers in his environment may need additional SAN storage or will run out of resources in the next 6 months”
This demo is about using the data in the DW to do predictions and forecasting. The MP ForecastReport.xml in the attached file can be imported and should work without any changes. When you run the report(Reporting > Forecast Report) you will see the parameters you need to fill in. The date time ranges (To and From) are the input data from what exists in your DW and the days to forecast is how many days of prediction you need.