One of the questions that have been extensively debated in the cloud management software industry is whether VMware would focus upon managing “its silo” (vSphere) and leave enterprise management and its complexity to other vendors (both large and small), or whether VMware would choose to try to become an enterprise cloud management software vendor themselves. That question has now been answered – at least in terms of VMware’s intentions.

VMware’s New Cloud Management Software Strategy

With its announcements at VMworld Europe today, VMware has unveiled a new management software strategy based upon the following principles and promises:

  • Comprehensive set of Management Offerings – Collectively, vCenter Operations Management, vCloud Automation Center, vFabric Application Management, and IT Business Management represent enough coverage of the waterfront to make VMware into a true management software vendor.
  • Cross Virtualization Platform Management – The core offerings listed directly above all either support or will support vSphere, Microsoft Hyper-V, Citrix XenServer and Red Hat KVM. This, along with the point directly below, is a huge change in VMware management software strategy.
  • Cross Cloud Management – The core offerings can manage across vCloud based clouds and either support or will support Amazon EC2 and Microsoft Azure as well.
  • Physical Systems Management – The core offerings can manage operating systems instances deployed on virtualization platforms or directly on physical hardware.
  • A new edition of vCenter Operations, Foundation, which will be bundled for free into every edition of vSphere.
  • Application Management Differentiation – Much of first generation and legacy systems management falls short because it fails to manage what the business really cares about, which are the applications that the users and the business rely upon. By starting the application management process with structured provisioning of the application in vFabric Application Director (which now works for Linux and Windows applications – both custom developed and purchased), vFabric Application Director abstracts the intelligence of how the application is constructed out of the application into a container that can then also specify how the application is monitored.

The State of the Management Software World

One of the things that has been so broken about management in the physical world is that many customers have hundreds of management tools and profess to hate most or all of them. This has come to pass for two reasons. The first is that legacy systems management vendors sold suites of functionality to executive management at the enterprise. Management bought these suites because they wanted one throat to choke on the management front. But things that looked like they were integrated on paper often ended up being Franken-Monitors, and as operation teams tried to deploy these suites and frameworks they found that various components either did not work, were too hard to use, or were missing key pieces of functionality.

Over time, as operating system and application platforms proliferated, point management solutions were brought in to address each new platform or environment. This led many companies to have different management products for Windows and Linux, different ones for Java and .NET, different ones for each database server, etc.

These problems were exacerbated by the new requirements that were created by virtualization and cloud computing. The software that worked for dedicated and static environments did not work in dynamic and shared environments. Cloud computing distributed workloads across organizational boundaries, something that legacy systems management vendors were completely unprepared to address.

Therefore, the current state of the management software world is that what most customers own directly impedes them from achieving the operational agility and operations expense savings that should fuel the next wave of virtualization and cloud computing. VMware has now stepped into this breach with a strong set of intentions to address these issues.

Implications for Enterprise Customers

One of our key recommendations for the last several years has been that there is nothing wrong with having more than one virtualization platform. There is nothing wrong with having more than one cloud. There is nothing wrong with having a system that is only partially virtualized, as some aspect of some systems still need physical hardware. But, there is something drastically wrong with building a separate management stack for each incarnation of virtualization platform, cloud, and hardware environment that you own. Enterprises should implement one and only one set of management software offerings that deal with all of the combinations of virtualization platforms, clouds, and hardware environments. VMware is now unique in that they are the first vendor to state the intent to deliver against these requirements, and they have started that delivery process with the offerings announced today. As enterprises think about how to manage the new world, they should think about what new management software to buy to replace the current mess that they own. VMware has certainly earned a place on that short list with this new strategy.

Holes in the VMware Offerings

The single most important issue for any enterprise evaluating VMware’s management offerings is this. Platform vendors (VMware with vSphere, Microsoft with Windows, Red Hat with Linux) tend to be excellent at evolving and promoting the success of their platforms. These same vendors tend to be mediocre at managing anything else other than their platforms. There is also the issue of trust. Most enterprise customers do not trust the fox to watch the hen-house  That means that most enterprise customers would prefer to procure a management solution for N platforms from one of the vendors who do not provide the platforms. There are good reasons for this – with the desire of letting vendor X watch vendors Y and Z at the top of the list. As with any suite suite of this magnitude there are also some important details to attend to:

  • Right now the vCloud Suites are more bundles than they are suites. The component products are not yet will integrated. There are still too many different virtual appliances, databases and management consoles. It will likely take VMware years to address this issue as integrating disparate products is always a complex and time consuming task.
  • Spending time integrating these products will make less time available to enhance their functionality. Therefore VMware risks becoming one of the legacy management vendors (slow to adapt to change) instead of being a vendor who disrupts the management software market.
  • Some of the new support for other virtualization platforms, clouds, and physical hardware relies upon agents. VMware has not been a strong proponents of agents in VM’s before – but now especially with the Enterprise Edition of vCenter Operations is reliant upon them for some critical functionality. This will give rise to the quesiton of how to manage all of these agents, a question for which Intigua has an intriguing answer.
  • Having vCenter Ops, vCloud Automation Center, vFabric Application Director and the IT Business Management products support multiple hypervisors, multiple clouds, and physical hardware is not quite the same thing as having a fully unified adminstration model across all of these environments. Right now vCenter still only manages vSphere and the native consoles for all of the other environments are still needed – unless you happen to choose Hotlink to address these issues.
  • While DynamicOps was an extremely strong and successful offering before VMware bought it, it was not without competition. Renaming it vCloud Automation Center and putting it in the high end of the vCloud Suite will not change how it stakes up against alternatives like Cloupia (who offers excellent integration with converged infrastructures), ManageIQ (who has built extensive enterprise class management functionality in their solution) and Virtustream (who is the only cloud management vendor to offer application response time based SLA’s).
  • While vFabric Application Director allows you to set auto-scaling rules for an application system, and DRS can certainly move VM’s around based upon resource constraints in hosts, there is still nothing that allows for workloads to be prioritized on the basis of service level like is possible with VMTurbo.
  • On the Operations Management front in the SMB/SME market, VMware faces three competitors, Splunk, ManageEngine, and SolarWinds who have built high velocity easy-to-try and easy-to-buy businesses with large numbers of customers. All three vendors offer free versions of their products. Therefore even though vC OPS Foundation is free, it is not without competition.
  • At the very high end of the Operations Management market, the Operations Management problem becomes a big data problem. 100,000 VM’s monitored every 20 seconds produce too much data arriving at too rapid a pace to be handled by monitoring solutions built on traditional databases. This is where Reflex Systems who attacked this problem with a clean sheet of paper is worth looking at.
  • At the APM layer VMware faces a multitude of competitors. For the segment of the market focused upon monitoring modern custom applications in production the high end of that market is dominated by Compuware/dynTrace who have significant integration with enterprise middleware that VMware lacks. In the new market for monitoring custom applications characterized by multiple platforms and scaled out architectures, vendors like AppDynamics and New Relic have significant market traction and support more applications platforms than vFabric APM does. In the new market for monitoring every application in production, new vendors like AppEnsure, AppFirst, BlueStripe, Boundary, Corrrelsense, and ExtraHop Networks offer greater breadth and deeper functionality than what the former AppSpeed product brings to vFabric APM.
  • Owning the structure of an application system is one thing. vFabic AD addresses the question of how an application system is constructed, but it does not address the question of how to model and manage system and application images across platforms. Thia is where scripts and legacy BladeLogic implementations are being replaced by Puppet, Chef, ScaleXtreme and rPath.
  • Other than certain components of the IT Business Management Suite, VMware’s management offerings are all on-premise. This made sense when all they managed was on-premise installations of vSphere or other hypervisors. But now the scope is multiple hypervisors and multiple clouds (in different locations, run by different companies). Managing across clouds invariably brings up the question of using a service to manage these distributed environments. Vendors like New Relic, AppFirst, AppDynamics, Boundary, Cloud Physics and AppEnsure all offer monitoring as a service – with the vendor hosting and maintaining the back end. This has the advantage of allowing the management agents in the various clouds to “phone home” which is easier to pull off than trying to poke holes in firewalls between various cloud providers and your enterprise.

Implications for the Management Vendor Ecosystem

Virtualization and cloud computing are disrupting the management software business in a way that it has never been disrupted before. This new disruption is being driven by three factors:

  1. Virtualization and cloud computing do not just create new requirements that need to be met, they also break legacy management solutions. It is simply not possible to manage a dynamic, shared and distributed system with the same tools that were used to manage a static, dedicated and locally isolated system. Many enterprises recognize this and uninstall legacy management tools when they virtualize servers.
  2. The way in which management software is bought and sold has been transformed. Taking the CIO and his staff out for a golf outing has been replaced by the “freemium” process of letting customers download the product and try it in production for free before asking for the order (with this being done over the phone and not via an on site sales call).
  3. The virtualization team at the enterprise has become a buyer of management solutions. This has allowed many new vendors to enter the market as they can sell to the new audience, one that is largely ignored by the legacy enterprise software vendors (IBM, CA, HP, BMC).

By employing a freemium model to build awareness for vC Operations and then to pursue an up-sell strategy to one of the paid editions (or a vCloud Suite) VMware has positioning itself to take full advantage of the above dynamics.  The first effect of this is going to be to put enormous pressure on the management software businesses of the legacy systems management software vendors. The message is going to be pretty simple. Once you get to 80% or 90% virtualization, no matter on what platform, “we (VMware) can take care of management for you, and you will not need those guys any more”.

What about the startup ecosystem? Here innovation rules the day. These companies exist because they tend to out-innovate their larger competitors. If you drill down into any one of the leading edge innovators in this market you will likely find things that their products can do that none of the larger companies (including VMware) can address. However, these companies tend not to offer full suites of functionality as VMware has now chosen to do. Therefore, it is now time for the members of the ecosystem who possess individual best-of-breed components to partner with vendors who have individual best-of-breed components at adjacent layers of functionality. How enterprises should approach this will be the subject of a separate post.

Summary

With this set of announcements, VMware has proven its intentions of joining the ranks of the major cloud management software vendors. Virtualization and cloud computing represent disruptions that create an opportunity for a new multi-billion dollar management vendor at the table. VMware has announced a suite of management functionality that makes it a credible vendor at this table. Legacy vendors of management solutions like IBM, CA, BMC and HP are now served notice that their businesses and customer footprints are at risk. Game on.

 

 

Share this Article:

Share Button
Bernd Harzog (336 Posts)

Bernd Harzog is the Analyst at The Virtualization Practice for Performance and Capacity Management and IT as a Service (Private Cloud).

Bernd is also the CEO and founder of APM Experts a company that provides strategic marketing services to vendors in the virtualization performance management, and application performance management markets.

Prior to these two companies, Bernd was the CEO of RTO Software, the VP Products at Netuitive, a General Manager at Xcellenet, and Research Director for Systems Software at Gartner Group. Bernd has an MBA in Marketing from the University of Chicago.

Connect with Bernd Harzog:


Related Posts:

Leave a Reply

Your email address will not be published. Required fields are marked *


three + = 10