Item: IBM Turbonomic
Rating: 9
Author: Kevin Stay

Use Cases and Deployment Scope

VMTurbo Operations Manager has been deployed in the AMER region and is scheduled for deployment to EMEA and APAC the coming fiscal year. It is to be our global standardized solution for the virtualization environment for analysis, planning, and automated remediation. Capacity planning or modeling the impact of migrations and major patching were nightmarish prior to VMTurbo. Rightsizing VMs simply did not take place. Identifying resource contention causing performance issues was also much more time consuming and far less accurate.

Pros and Cons

The highly customizable reporting capabilities and automated delivery of reports to application owners and server administrators has no equal in any other product we considered.
The consumer model used as the basis for the entire product is unique in the industry. It provides insight across the entire stack not available in products pulling everything exclusively from vcenter.
In particular classifying storage volumes as having a specific capacity including oversubscription rate (due to dedupe, compression etc.), IOPs and latency and then tracking the use of that total capacity in each area and recommending remediation based on that is extremely valuable.

Custom settings per individual volume for total capable IOPs and oversubscription rate are not available at least in the standard enterprise version without add-ons. That can only be done at the VMware cluster wide level.
The interface is unfortunately in flash. The sooner it moves to pure HTML5 the better.
The standard reports included are outstanding. Including a similar set of standard dashboard user could then use as examples or customize would be a great addition.

Return on Investment

The amount of time required to identify the root cause of performance issues has been cut by an order of magnitude. In addition a number of those issues are now avoided ahead of time altogether by following preventive recommendations.
The ability to accurately plan capacity purchases based on actual historical utilization growth curves is fantastic. Previously it was always just ask for as much capital as you can each year and spend it all. Massive over provisioning wasted a LOT of money. Also, the ability to manually add in coming business projects and required number of servers of size X as part of those models makes for much more accurate forecasts then just using past utilization.
Being able to execute recommended remediation on a VM (host migration, volume migration, add/remove RAM or vCPU) with a single click from within the tool also saves a lot of time along with seeing not only what the current environment state is, but also what it will be post remediation. Fantastic.

Alternatives Considered

vCOPS, Veeam ONE and Fogworks

vCOPS is a mess. The interface is kludgy and counter-intuitive. It has no ability to query or interact directly with storage systems or network switches; everything is from vcenter. Reporting is weak at best.

Veeam ONE would probably be OK for smaller uses with customers who are already long term committed to the main product. Like every other alternative considered it does not classify volumes as having a certain IOPs capacity, but simply reports and current and historical utilization.

Fogworks came the closest to VMTurbo in our evaluations. In the end the ability to fully automate the parts of remediation you choose and classifying volumes as having a specific IOPs capacity won the day despite the formers interface being superior.

Likelihood to Recommend

In a properly designed and implemented VDI environment supporting a small range of use cases VMTurbo Operations Manager would be of fairly limited value. Smaller installations with fewer than three hosts or a couple dozen VMs would also see limited benefit. Even organizations with large data centers managed by VMTurbo Operations Manager would get little from also using it to manage their smaller sites of those sizes.

VMTurbo Operations Manager is especially strong in larger sites of several to several dozen hosting servers and hundreds to thousands of server OS VMs. The efficiencies you will gain and the accuracy of the growth planning will save you a lot of capital while still assuring expected performance levels.

Products Replaced

Key Differentiators

Product Features
Third-party Reviews

The powerful utilization types of custom reporting and recommendations made possible by the whole stack "consumer model" is unique. Same thing for volumes having IOPs "capacity" that is consumed upstream by a certain percentage of what can be reasonably expected from them. No considered alternative used such a model or had such capabilities.

Evaluation Lessons Learned

I would not have wasted so much time with other evaluations of products that did not even claim to have required capabilities.

Implementation Rating

There is value to be had in just doing a "click install" and then just evaluating and following recommendations. However, that is nowhere near the benefit you will get from the very powerful custom reporting if someone is willing to take the time to master this platform that can be somewhat demanding at that level.

Implementation Details / Implementation Partner

Implemented in-house

Implementation Phases

Yes - Salt Lake City datacenter in AMER region was the pilot and first implemented in production. Palo Alto main global datacenter then implemented on new VMware clusters and workload migrated there. EMEA and APAC scheduled for FY16. Functional phases also used by starting with monitoring and reporting and then adding limited remediation automation one step at a time.

Change Management Lessons

Change management was minimal - If you do not have strong institutional change management in place VMTurbo will not magically create it for you.

Implementation Issues

Ensuring access permissions were set correctly in VMTurbo Operations Manager and in all the environment to be managed.
Implementing signed certs could have been far more automated as is the case with Dell iDRAC for example.
The learning curve, especially for your primary administrators, is not particularly shallow. Devote a significant amount of time and effort to understanding this powerful toolkit if you expect to get full value from it.

Support Rating

Just this week we had an issue where security permissions we had done wrong were causing problems in VMTurbo Operations Manager. Despite this being our fault the service request was handled very quickly. In addition we are routinely contacted proactively to see if there is anything we need or they could do for us.

IBM Turbonomic Customer Support Pros and Cons

Pros	Cons
Quick Resolution Good followup Knowledgeable team Problems get solved Kept well informed No escalation required Support understands my problem Support cares about my success Quick Initial Response	None

Premium Support

Yes - We now depend of VMTurbo to assure ongoing performance. The small price of paid support is worth every penny.

VMTurbo, just the operations performance boost you needed.

Software Version

Modules Used

Overall Satisfaction with VMTurbo Operations Manager