2011 Winners and Losers in Virtualization Management

The virtualization management industry (all aspects of physical systems monitoring, virtual systems monitoring, applications monitoring, and systems/applications provisioning) is undergoing an extremely rapid rate of change. In 2011, VMware put two very important stakes in the ground. The first was to combine performance management, capacity management, configuration management with self-learning analytics in vCenter Operations. The second was to state that its forward going management strategy was around delivering OPEX savings through automated remediation (monitor–>remediate–>notify). These VMware initiatives will shape how management plays out in 2012, and create new winners and losers.

Virtualization Management – Not Winners in 2011

One of clear trends in systems and performance management is that customers do not want to repeat the mistakes made in managing their physical environments when it comes to managing their virtual environments and private clouds. This has resulted in customers moving away from products that are expensive to purchase, expensive to own, and require large amounts of labor (typically vendor provided consulting) to implement and keep running.

The second factor that is creating a clear set of losers is that the correct way to manage a dynamic and shared virtualized system is very different than the way in which physical systems have been managed in the past. Specifically, most of the management of physical systems has been done by collecting utilization information from those systems and then inferring the performance of the workloads running on those systems by looking at normal vs abnormal levels of resource utilization. Inferring workload performance from resource utilization statistics does not work in virtualized environments primarily due to the fact that resources are too shared and abstracted to be a useful indicator of workload performance.

Therefore the set of losers in the Virtualization Management industry for 2012 are vendors whose products are expensive to buy and own, and who have not adapted their products for the new requirements of virtualization and cloud. And the losers are the legacy systems management vendors like IBM, CA, HP and BMC.

Who is Who in the Virtualization Management Ecosystem?

Before we get into specific winners and losers by category, let’s have a look at the entire landscape. The diagram below shows the prominent vendors in the virtualization management ecosystem (leaving out backup and security which are covered separately on this site).

Virtualization Management Vendors2
Operations Management Winners in 2011

The vast majority of the customers who run server virtualization in production are still just using the basic management capabilities of the management console for the virtualization platform. For example less than 20% of VMware vSphere customers use anything else other an vCenter to manage their vSphere environment. This clearly has to change as the workloads running on the vSphere platform go from being primarily low hanging fruit applications that IT owns (and can virtualize without asking anyone’s permission) to virtualizing business critical and performance critical applications. The winners here are:

  • VMware – for combining resource utilization based performance management and capacity management with configuration management, and bubbling all of the metrics up into Health, Efficiency, and Risk scores with self-learning analytics in vCenter Operations. VMware further deserves credit for putting the automated remediation stake in the ground as its forward going strategy with its management offerings. This is crucial step forward for the virtualization industry, as it creates a new hard dollar ROI for virtualization – OPEX savings from IT Automation. This new ROI is crucial as consolidation ratios are not going to be as high for the next 60% of the workloads to be virtualized as they were for the 40% of the workloads that have been virtualized.
  • Zenoss – for delivering a comprehensive availability management framework for physical and virtual systems that can fully replace the legacy solutions from IBM, HP, BMC and CA without imposing the costs and complexities of these legacy products upon next generation virtualized systems and the hardware that supports them.
  • VMTurbo – for being the only vendor that is delivering automated workload service assurance today on the vSphere platform. In other words, VMTurbo is delivering today, what is promised in VMware’s future strategy.
  • Veeam – for successfully tying vSphere into Microsoft SCOM via its nworks for Microsoft SCOM product. This is one of the few solutions that allows you to use one comprehensive management solution (SCOM) to manage two virtualization platforms (vSphere and Hyper-V). This avoids the critical mistake of building a management stack for each virtualization platform which is one of the mistakes that plagues management on the legacy physical side.
  • Reflex Systems – for adding Capacity Management to their already extremely functional and scaleable  virtualization management platform.
  • Quest Software – for acquiring vKernel and therefore being the first vendor in the ecosystem to react to the suite of functionality that VMware is offering.
  • Hotlink – for shipping their revolutionary “Supervisor” that allows you to use VMware vCenter (and any other product that relies upon the vCenter API’s) to manage vSphere, Microsoft Hyper-v, Red Hat KVM, and Citrix XenServer.

Infrastructure Performance Management Winners in 2011

Infrastructure Performance Management is a separate category of vendors from Operations Management because these vendors address a crucial question for virtualized systems. If it is not possible to infer the performance of a virtualized system by looking at how resources are being used, then how does one do it? The answer is by looking at how long it takes to do things, specifically at the latency for requests for work placed upon the infrastructure by its workloads. Winners are:

  • Virtual Instruments – for being the only vendor that can tell  you the latency of your storage I/O requests in a real-time, continuous, deterministic, and comprehensive manner. VI does this by tapping the SAN and seeing every storage transaction that flows through the SAN to and from the storage arrays.
  • Xangati – by combining the storage latency information from vSphere with its own view of network latency and performance, yielding a solution that is particularly useful in XenDesktop and VMware view deployments. Xangati’s information is so useful in these situations since desktop virtualization puts both the storage subsystems and any wide area networks that serve users under enormous stress.
  • Confio Software – for being the only vendor that can tell you the latency of the requests made to your databases, in combination with the latency of the requests produced by the databases to the supporting storage subsystems.

Application Performance Management Winners in 2011

  • VMware – for announcing a product, vFabric APM which is the first product to calculate a Performance Index for every monitored application. The PI is a normalized score between 0 and 100 (100 is good) which is based upon both response time itself and variability in response time. This will allow for the direct comparison of performance across applications, and will ultimately allow for their performance to be both prioritized, and for the performance of the important applications to be automatically guaranteed through service assurance.
  • AppDynamics – for producing an APM solution that is truly easy to install and get working, and for pushing the functionality envelope with automatic business service discovery and workflow automation features for Java and .NET applications.
  • New Relic – for being the first APM vendor to succeed with a SaaS model, and for being the first APM vendor to support Java, .NET, PHP, Ruby, and Python based applications.
  • BlueStripe – for being the first APM vendor to be able to discover application topologies in a completed automated manner, and then measure hop-by-hop and end to end response time for every Windows, Linux, AIX, or Sun OS application whether it is purchased or custom developed, and whether it runs on a physical or virtual infrastructure.
  • Compuware – for acquiring dynaTrace and embarking upon the path to integrate it fully into the Gomez product line – ultimately producing one of the richest suites of APM functionality on the market.

Cloud Management Winners in 2011

Cloud Management came into its own in 2011, due to VMware placing a great deal of focus upon the space with vCloud Director. The key event during the year was Cisco acquiring newScale, the leading independent vendor of service catalog frameworks. Platform Computing, Gale Technologies, DynamicOps and Abiquo are all winners this year for successfully defining the cloud management space as a cross-hypervisor problem – not one that should be addressed with a different cloud management stack for each virtualization platform (hypervisor) in the enterprise.  Platform Computing, Abiquo, and Gale Technologies are winners for starting the process of being able to ensure the performance of workloads that are dynamically provisioned in private clouds by allowing high priority workloads to have guaranteed resources assigned to them. This will ultimately lead to the marriage of the service assurance and private cloud management businesses, as business critical applications are not going to be provisioned via cloud management stacks until service levels can be assured. Finally Embotics is a winner for pioneering the “cloud in a box” idea – meaning that you can get a private cloud up and running quickly and cheaply.

Image Provisioning Winners in 2011

This is a new category that just came into focus in 2011, so every vendor in this category (Puppet, Opscode, ScaleXtreme, and rPath) is a winner simply for being there at the inception of the revolution of yet another aspect of systems management. The bottom line is that images are not the right fundamental object that should be managed in a virtualized or cloud based environment. The level of granularity needs to be much deeper, in order to ensure that components of images can be individually updated and rolled back and so that many images can be built on the fly across deployment platforms in a rapid, automated, and managed fashion.


The management ecosystem for virtualization started to transform significantly in 2011, driven by VMware’s new management strategy and management offerings. The big four are now boxed into an untenable position with expensive software that is hard to buy and hard to deploy. In 2012 there will be aggressive partnering in the ecosystem as vendors try to compete with the VMware suite by integrating with other vendors who have adjacent functionality.