By Porus Homi Havewala
We introduced the concept of Consolidation Planning in the previous parts of this article series. Part IV was here. This is Part V.
Click on the Destinations tab. This is displayed in Figure 1-21.
Figure 1-21: Destinations Tab.
In this tab, we see that the Destination Server is the phantom Exadata Database machine with four rack nodes (since it is a quarter-rack). The SPECint rate is shown to be 391 (estimated) for each node. The CPU and Memory utilization that would result after the consolidation on the two nodes, are also shown.
We can see that the one of the rack nodes would reach 64.2% of CPU utilization, which is more than the other nodes. Click on the Ratio tab. This is displayed in Figure 1-22.
Figure 1-22: Consolidation Ratio.
In this tab, we can see the Consolidation Ratio of Source servers to the Target servers calculated as 1.3. This is because only 5 Source servers have been consolidated to the 4 rack nodes of the database machine. Click on the Mapping tab, as seen in Figure 1-23.
Figure 1-23: Mapping Tab.
On the Mapping tab, the destination and source servers are shown together along with their SPECint rates and the resource utilizations. This allows you to get an idea of how the CPU and Memory utilization of the destination servers has been calculated. The Destination Mapping is displayed as Automatic.
Click on the Confidence tab. This is now displayed in Figure 1-24.
Figure 1-24: Confidence Tab.
On this tab, the Confidence is shown as 100%. Out of the data collected from the source servers, 199 total data points were evaluated and all of them met the requirements.
The Heat Map graph shows the Hourly Resource Utilization of all resources (including CPU and Memory) for all destination servers, and is color-coded to show the hours and days when there is maximum and least resource utilization.
For example, in this particular scenario, there are blue colored blocks showing 0-20% utilization on certain days and hours. This seems fine.
You can click on any of the blocks in the graph to see the value. Also, using the drop down box for the destination server, it is possible to choose any one of the rack nodes, and see the utilization for that node only.
Move to the Violations tab. There are no Violations to be seen in this scenario. Next, move to the Exclusions tab, shown in Figure 1-25.
Figure 1-25: Exclusions Tab.
There are 4 server exclusions visible on this tab, these servers could not be consolidated, since their memory requirement at a certain hour was higher than what was available.
So, we obviously require a more powerful Exadata machine.
Go back and create a new what-if scenario. In this case, select an X5-2 (Eighth Rack). This is a more powerful machine. Also remove the constraints. The scenario analysis completes as seen in Figure 1-26.
Figure 1-26. The New Custom Scenario for an Eight Rack.
Click on the Ratio tab. We can see that all 9 servers have now been consolidated on the two nodes, 3 servers on the first node and 6 servers on the second node. (Figure 1-27).
Figure 1-27. The Ratio Tab
Move to the Mapping tab (Figure 1-28). The Rack nodes in this case have a higher CPU Spec metric of 700 (Estimated), and the memory is also higher at 256GB per node. So all of the 9 servers could be consolidated.
Figure 1-28. The Mapping Tab
The confidence tab also shows 100% confidence (Figure 1-29), although there are hot spots on some hours approaching 90% usage.
Figure 1-29. The Confidence Tab
There are no violations and exceptions.
In this way, we have seen the first P2P (Physical to Physical) project and scenario results, with their what-if scenarios.
(1)Consolidating to Oracle Public Cloud Servers
Enterprise Manager 13c’s Host Consolidation Planner also allows you to select the Oracle Compute Cloud, as one of the possible phantom destination server choices. In Figure 1-30, we have created a new scenario and selected to use the Oracle Compute Cloud.
Figure 1-30. Consolidating to the Oracle Compute Cloud
The list of configurations available for the Oracle Compute Cloud is seen in Figure 1-31. We have selected the OC3M configuration. This is a “compute shape” available in the Oracle Public Cloud with 4 OCPUs and 60 GB RAM. Refer to the Oracle public cloud link https://cloud.oracle.com/compute?tabname=PricingInfo
Figure 1-31. Selecting the Oracle Compute Cloud Configuration.
Click on Ok, and go through the remaining steps and submit the new OPC scenario. The analysis of the scenario completes successfully in some time. Select the completed OPC scenario. In the lower half of the screen, move to the Ratio tab.
Figure 1-32. Ratio Tab of OPC scenario.
In this Ratio, we see that 5 servers have been consolidated on to 2 target servers, with a consolidation ratio of 2.5. Move to the Exclusions tab.
Figure 1-33. Exclusions Tab of OPC scenario.
In the Exclusions Tab, we can see that 4 of the source servers have been excluded from the consolidation. The reasons are shown. For these servers, either the CPU requirement or the Memory requirement at certain hours were more than could be handled by the 2 target servers. So, they were excluded.
We continue this article series in Part VI which is here.
(This article is an excerpt from Chapter I of the new book by the author titled “Oracle Database Cloud Cookbook with Oracle Enterprise Manager 13c Cloud Control” published by Oracle Press in August 2016. For more information on the book, see here.)