This is part of a series of articles looking at real operational situations and how virtualization management solutions react.
You did it. You got the new hardware approved and delivered (no easy task in today’s tight economic conditions). Now, once it’s installed and connected—whether it’s one host or a whole new cluster—you have to decide what to do with your workloads. Does your management system help you make these important decisions? Let’s take a look.
For tools that rely upon trended history or “learning” your environment, there are a few gotchas:
- First, your new hardware may have different capabilities or resource configurations than incumbent equipment. Therefore, there are no trends on “day one” and no “normal” behavior to use as a baseline. So, you get no help in making decisions.
- Second, you will need to re-examine all of your thresholds since they may not apply to the new hardware. The state of your environment is “abnormal” in the eyes of your virtualization management solution. How long does it need to build a new baseline and how reliable is the alerting during this time. You will either get many “abnormal” alerts or no alerts at all since it has no basis to determine abnormalities while it’s in flux.
- Third, how do you manage performance and capacity during the transition to the new hardware? If you transition quickly to the new hardware you run the risk of creating bottlenecks since you don’t have reliable alerting (see point two, above). If you transition slowly, the system is continuously learning the new “normal” because of the constant change. All conditions question the reliability in the alerts you get or don’t receive.
So, what value do these types of tools provide when you need them most?
Here is how VMTurbo Operations Manager controls the situation to deliver the desired results:
- First, VMTurbo offers powerful planning features to simulate the additional hardware in your infrastructure with existing or additional workloads—prior to installation—as well as the decommissioning of old hardware. “What if” scenarios provide insight into the outcomes—not just a collection of data that you are left to analyze.
- Second, when you introduce the hardware into the estate, the compute resources of newly introduced hardware are added to the existing pool of compute resources available to workloads in your environment. Operations Manager models how VMs/workloads can take advantage of these resources and provides the specific control actions (and can automate the actions) to assure performance while making the most efficient use of resources (VMTurbo’s Software-Defined Control).
- Lastly, almost immediately, a “To Do” list of recommendations guides you in not only incorporating the new hardware, but in the ongoing (after installation of your new hardware) control of the environment to maintain your desired or healthy state.
VMTurbo enables you to take advantage of your new equipment sooner and better—with less risk. Sounds simple to me.
See related blogs: