The tour started on March 25 in St. Louis, MO and ends on November 28 in Chicago, IL, and it stops at 10 cities in the U.S. It's likely that the VMware Express will park up in San Francisco, Calif. in September in time for the annual VMworld event. But does VMware really need a juggernaut to get the message across about its virtualization platform?
Last week, I sat in on a Microsoft Virtualization Fast-Track course in an effort to be a more well-rounded "virtualization guy" rather than being a "VMware guy." In the short time I was there, however, it's clear that folks in Redmond have some mileage to make up for if they want their Hyper-V product to be remotely near the quality of the VMware's hypervisor.
I intend to write a detailed report both about the course and product. My current impression, however, is that Microsoft Hyper-V and System Center Virtual Machine Manager are two generations behind VMware products – and there were technologies in ESX 2.x and vCenter 1.x that
How does Microsoft threaten VMware?
So what's the threat? Well, I think the threat is not so much Microsoft claiming like-for-like feature parity with VMware. That would be a very hard position to maintain. The strongest position that Microsoft could play would be to claim that they offer "good enough" virtualization – virtualization ready for the vast majority of folks who haven't yet virtualized.
Clearly, VMware needs to get its message across that good enough virtualization is not, actually, good enough (nor is it good corporate marketing) – and put the squeeze on Microsoft's PR/marketing juggernaut. That's the message the VMware Express is meant to get across. Additionally, although many are aware that Microsoft has announced increased cooperation with Citrix and Hewlett-Packard, VMware has meanwhile quietly squeezed a few noteworthy partners on to its Express truck, including AMD, Cisco, Dell, Xsigo and NetApp.
Flashback to data centers-in-a-container
The whole setup reminds me of that series of "data center in a container" news articles in recent years, where a flurry of original equipment manufacturers (OEMs), such as IBM and Sun, demonstrated how they could pack a standard container, which you see on the back of trucks and container ships, and it could contain everything you needed to build a data center – servers, storage and network.
The data center in a container reminds me of some of the early laptops I used, where you did need a truck to move them about. All joking aside, the whole idea of the data center in the container is that the container has become the defacto unit of shipping globally and could become the building blocks of a data center. This concept is something all the OEM vendors are jumping on -- as witnessed by EMC's vBlock offering; so perhaps when you need a new data center it will arrive looking like the VMware Express Truck (hopefully not wrapped in a ton of cardboard and polystyrene).
VMware acquires... and acquires some more
Not content with touring the country by itself, VMware decided to acquire EMC. (...That's a joke, by the way. Not a week seems to go buy without some rumor rippling through the blogosphere about Company X buying Company Y.)
The truth is, VMware's deal to buy a collection of management products from its parent company has probably been in the pipeline for sometime. It seems like VMware is trying to rapidly evolve out of the narrow tag of being "the virtualization company" to something else entirely.
VMware has a development platform in the shape of SpringSource and an email application in the shape of Zimbra. Now VMware has coupled a raft of products that allow for additional management beyond the virtual world. My take on this is that VMware wants to appease critics who argue that VMware is a one trick pony, and only Company Y offers true end-to-end management in all areas of the stack. It also seems clear that EMC felt it was not in the best position to promote these products, which makes you wonder why EMC bought or developed them in the first place.
EMC's original ownership will eventually be forgotten as these products become known as VMware technologies; it does seem like these technologies seem to fit more neatly into the VMware agenda, anyway. VMware has also got its paws on the ConfigureSoft product, which allows for both ESX host and guest operating system configuration -- it comes with a vCenter plug-in to handle this -- and is also specifically aimed as compliance issues. I figure this system will find a natural compatibility with VMware's Host Profile feature and ConfigControl auditing system.
VMware also acquired the FastScale product, which is an engine for virtual machine (VM) creation. I think VMware may want to integrate this with its Linked Clones"feature of VMware View. Finally, VMware acquired nLayers, which is an agentless discover system that maps out your network and the relationship between VMs and physical machines. It's not 100% clear where VMware will place this technology, but I see it as a natural extension of the existing "map" views within vCenter.
Wrapping it up: VMware's strategy
It's seems clear that VMware is taking a dual strategy in efforts to grow outside of its remit. Where it can be done (or should I say done better by?) VMware engineers, VMware seems to be developing technologies in-house (such as VMware Site Recovery Manager), in other cases where VMware can leverage a company or technology it can license or buy of the peg – it will do that.
The worry that some commentators have is that this buy-in approach might leave the vSphere 4 suite feeling like a collection of products that have merely had their logos changed, rather than being fully and seamlessly integrated. I've seen this happen when VMware acquired the Propero Workspace product, which became VMware Desktop Manager (VDM) and then morphed into VMware View. It does take time to make an acquisition look and feel like it's something that was designed from conception in-house.
The real struggle playing out here is a scramble by various vendors to demonstrate to customers that they have a complete range of tools capable of managing the stack enabling the cloud to function. VMware is not alone in its need to fill various holes and limitations in managing that stack. It may be the case that merely having an offering and "checking the boxes" is what counts in the short term. It seems like a smart move to not waste valuable development dollars on technology your parent company already owns.
Mike Laverick (VCP) has been involved with the VMware community since 2003. Laverick is a VMware forum moderator and member of the London VMware User Group Steering Committee. Laverick is the owner and author of the virtualization website and blog RTFM Education, where he publishes free guides and utilities aimed at VMware ESX/VirtualCenter users, and has recently joined SearchVMware.com as an Editor at Large. In 2009, Laverick received the VMware vExpert award and helped found the Irish and Scottish VMware user groups. Laverick has had books published on VMware Virtual Infrastructure 3, VMware vSphere4 and VMware Site Recovery Manager.
This was first published in April 2010