We've been virtualizing our servers now for more than a decade... well, some of us have anyway.  But what we are finding as we virtualize more and more of these servers is that it becomes difficult to monitor and manage them, and traditional management applications just don't cut it.  And even the tools provided by the virtualization vendors themselves have gaps with missing technology.

That's where companies like VMTurbo come into play.  The company's founders have spent over two decades building data center management software, with the founding team coming from SMARTS, the company acquired by EMC and then later passed over to VMware to try and help solve virtualization management problems.  The group formed VMTurbo back in 2009, and they are building purpose-built virtualization management software to address those technology gaps left by the server virtualization platform vendors in a new, non-traditional way.

I was able to catch up with Ilya Mirman, VP of Marketing at VMTurbo to find out a bit more about the company and their solution.  Here is part of our conversation.

VMblog:  What is the problem that VMTurbo is solving, and if you could, explain to us why this problem is so hard to solve.

Ilya Mirman:  Virtualized data centers need to allocate resources among virtual machines (vms) to best utilize resources; and to assure the qos of mission critical apps.  These tasks are too complex for manual administration because:

1. VMs' resource needs may change dynamically
2. There are complex constraints on which resources can service which VMs
3. VMs sharing a resourcevmay interfere with each other, disrupting app performance
4. This is a large scale optimization problem
5. Even greatly simplified versions of the allocation problems lead to unmanageable algorithmic complexity...

This complexity leads to great inefficiencies of virtualized data centers.  For example, under-utilizing most of the resources most of the time... and difficult performance problems of virtualized mission critical apps.

VMTurbo provides automated tools for the virtualized data center to optimize resource allocations and assure the performance of virtualized mission critical apps.

VMblog:  Can you talk about the needs of the dynamic data center and "the cloud?”

Mirman:  Managers of virtualized data centers and cloud infrastructures need to:

1. Optimize their cloud capacity to handle the aggregate workloads of apps;
2. Allocate resources efficiently to meet dynamic demands of apps;
3. Assure the performance of apps;
4. Enable users to budget the resources allocated to their apps’ business importance.

VMblog:  Can you tell us a bit more about the VMTurbo virtualization management suite?

Mirman:  VMTurbo delivers an Intelligent Workload Management solution for Cloud & Virtualized environments.  VMTurbo uses an Economic Scheduling engine to dynamically adjust resource allocation to meet business goals.  Using VMTurbo, our customers ensure that applications get the resources they need to operate reliably, whilst utilizing infrastructure & human resources in the most efficient way.

VMblog:  What is VMTurbo’s approach to solving some of these problems being faced in the virtualization market?

Mirman:  VMTurbo recasts the resource and performance management problems as balancing the supply and demand for resources.  For example, bottlenecks are formed when local workload demands exceed the local supply of resource capacity. This suggests the use of economic solution techniques to efficiently redistribute the demand, or increase the supply.  VMTurbo resource and performance management technologies are based on an economic model involving two sets of abstractions:

1. Modeling the virtualized IT stack as a service supply chain, where components, e.g., VMs, consume services of other components, e.g., physical hosts, and offer services to their consumers, e.g., guest OS’s and applications;
2. Using virtual currency to balance the supply and demand of services along this supply chain. Resource services may be priced to reflect imbalances between supply and demand, and drive resource allocation decision. For example, a bottleneck, reflecting excess demand over supply, will result in rising prices of the respective resource. Applications competing over the resource may shift their workloads to alternate resources to lower their costs, resolving the bottleneck.

VMblog:  So what are the user benefits?  And what is the ROI?

Mirman:  Sure, those are:

1. Optimize the use of vdc resources by virtualize apps, resulting in a 30-40% reduction in the virtualized data center CapEx
2. Automate proactive detection/diagnosis/handling of performance problems drives:
   (a) A reduction in the costs of “down-time” and “slow-time”;
   (b) A reduction in the labor costs of manual detection/diagnosis/handling performance problems;
   (c) Improved satisfaction among application owners / customers;
3. Assuring the performance of virtualized apps enables:
   (a) Virtualizing of mission critical apps and thus reduce their CapEx/OpEx
   (b) Improving performance of mission critical apps (and productivity of respective business units)
   (c) Properly pricing of performance guarantees
   (d) Improving data center customer satisfaction

VMblog:  And how does VMTurbo align IT with the business needs?

Mirman:  Virtualized data center and cloud workloads fluctuate -- by season, by quarter, by the day of the month, day of the week, time of the day, by the second.  At the same time, workload priorities fluctuate -- by application, by business, by time, by user.  Infrastructure managers need AGILITY.  Cloud IT environments are increasingly shared, dynamic, constrained, and growing in complexity. Cloud IT environments need AGILITY.

Today’s IT management is struggling to meet these challenges. Yesterday’s tools fall short: Silo’d by technology and function, stuck in the weeds of collecting too much detailed data, swamped by a myriad of point tools with little intelligence to automate decision making, IT management is unable to adapt to the demand of the new environments.  Yesterday’s tools are unable to orchestrate across multiple layers of services and infrastructure, leaving the heavy lifting to system administrators.  Invariably, ad-hoc approaches are used to address exceedingly complex problems.  Large, complex, dynamic environments are manually managed, yielding performance degradation, inefficiencies, waste and unscalable operations – failing to deliver the agility promised by virtualization and cloud computing… Enter VMTurbo.  Our customers ensure that applications get the resources they need to operate reliably, whilst utilizing infrastructure & human resources in the most efficient way.

VMblog:  In your opinion, what's the approach that other vendors take to address the problem of virtualization management?

Mirman:  Alternative approaches involve threshold-based systems based on programmed rules (“if a metric is X, do Y” -  e.g., “if vMemory or vCPU are below a predefined threshold then re-size” or “if metrics are outside of normal boundaries of operation raise an alert”). These types of systems do not work when 10s, 100s or 1000s of different variables – across multiple physical and virtual resources –  need to be considered holistically in the decision-making to determine what ACTIONS to take.

VMblog:  So what then is unique about the VMTurbo suite?  And how is this different than the other approaches? 

Mirman:  The value of virtualization management systems depends on their underlying technology strength.  VMTurbo technology is based on transforming the virtualized resources into a marketplace where VMs shop competitively for their workload needs, using budgets reflecting the business significance of their apps.  This model provides a unified systemic solution to all resource and performance management needs.   Other tools are based on a patchwork of ad-hoc solutions for each resource/performance management problem.

###

I'd like to once again thank Ilya Mirman, VP of Marketing at VMTurbo for taking time out to speak with us today and answer a few questions.