VMware, MS Hang Back as Others Catch Up on Virtualization
- 08 August, 2008 13:58
There are two ways to look at the rush of secondary and tertiary products coming into the virtualization market-especially this week and next week.
Symantec, for example, bought nSuite to give its products a greater ability to ask for and manage connections between servers, virtual servers and the various types of endpoints that connect to either one. Israeli remote-access software developer Ceedo announced a product designed to let end users connect to Citrix app servers securely, even without a pre-installed client on their own machines.
Citrix may raise its prices. A UK company is putting a virtualized data center into the cloud. OpenCloud Computing has announced its OpenNebula Virtual Infrastructure Engine, which appears to be an attempt to virtualize a data center, rather than just a server.
After years of supporting VMware, VMLogix announced its LabManager VM lab-management software would support Hyper-V as well as VMware.
Backup vendor Veeam announced a V-specific version of its products.
And Virtual Iron is getting ready to announce (Monday) LivePower-a new feature for its virtualization software designed to automatically consolidate VMs into hosts with spare capacity in order to use as few servers and as little electricity as possible.
(This is, of course, to completely ignore, for once, the continuing Microsoft-VMware spitting match which, this week, is over which vendor's stuff is cheaper, the inexpensive Microsoft software, or new, inexpensive version of VMware's.)
The first is that virtualization is the hottest fad in computing and every vendor with technology that can be made relevant on even the most shallow level will past a "Now Virtualized" sticker on the box and try to get in on the frenzy.
That's not only the most obvious interpretation, it's also the most defensible historically. There's plenty of innovation in the computer business, but there's even more "innovation," that the "innovators" compare to Levi and Strauss selling gear to miners, but is actually more like coopers and wheelwrights following the pioneers West to do exactly the same thing they'd always done, but with poorer facilities, fewer tools and at higher prices.
There are a million real-IT examples, but the most ridiculous is actually the iPhone (and iPod, actually)-a genuine innovation that spawned a million new packages for headsets and keyboards and chargers and headsets that are exactly the same as the ones already on the market except for the annoyingly proprietary add-ons Commisar Jobs requires for all Apple hardware, lest it be mistaken for anything pedestrian, easy to use or cost-effective. (Why, for an example to the example, do iPods still use proprietary hardware/power interfaces when accessories ranging from keyboard mini-fans to MP3s to terabyte-scale portable storage devices do just fine with standard USB A or B plugs?)
Fortunately for virtualization customers, though not for the vendors, the second way to interpret the latest wave of virtualization-tagged product announcements is a lot more useful:
It's not that vendors are piling on the bandwagon, it's that the bandwagon is growing large enough that they have no choice but to jump on.
That's good because there will be a lot more management, security and other secondary products available in the near term than there would be if virtualization remained a niche function, or if the technology remained so esoteric that add-on products really did have to be customized and tweaked to support virtualization as they did to support clustering, grid computing, broadband and wireless networking and other leaps forward.
Virtualization, as I am far from the only one to point out, has become a commodity function, or is rapidly becoming one. That doesn't mean that high-performing virtualization software has no value above that of more basic approaches. It means that every time a company of any size (which usually mean biggish companies, but in virtualization is quickly coming to mean companies of any size) is either installing virtualization software, or plans to do so.
Yankee Groups's latest report on the virulence and spread of virtualization estimates that 75 per cent of end user companies are either moving forward or are preparing to move forward with virtualization installs.
That's not a fad. It's not even a movement. It's the dominant approach to server-farm management.
It doesn't mean every server purchase involves virtualization products, or even that every server purchase will spark a discussion about virtualization.
It does mean that any vendor selling anything designed to operate inside a data center, connect to a data center, or allow the words "data center" to pass through its data bus at any point during its lifecycle had better have a response when customers ask "how's this going to work with my virtualization setup?"
They'd better have a good answer even if the customer doesn't know what that virtualization setup will be. Virtualization is so much an assumption on the part of customers that it is no longer the responsibility of the customer to explain that project ABC involves some virtualized elements.
It's up to vendors pitching for that business, consultants designing the software or infrastructure and anyone else involved to assume there will be some storage or server virtualization involved and take that into account, rather than pitching a non-VM-capable plan as representing their basic expectations and offering a VM-able version for an additional cost.
That's good for customers who are looking for good add-on tools for security or management or integration with various applications or non-virtualizable systems.
It's not a guarantee any or all of those vendors will actually know what they're talking about when they talk virtualization; customers will still have to vet what the vendors or integrators are telling them so as not to sign on with a company with more buzz-awarness than technical skill in that particular area.
It's not particularly good news for vendors whose products don't either supply or directly support virtualization. Those tertiary vendors have to compete for attention, just as they always have, while rushing to add demonstrable, practical functions that plug weaknesses peculiar to virtual servers rather than the physical kind.
That's what's going on now-hundreds of vendors are rushing to get into Virtualization Hall before customers start wondering why they weren't there already.
The most interesting physical manifestation of this rush is going to be at the US Fall trade shows-especially VMware's 14,000-attendee extravaganza at the Venetian in Las Vegas September 16-18.
In-person conferences and trade shows have become a lot less important to both customers and vendors over the last few years-largely because of the increasing volume of product and technical data available free online, the success of virtual events, local events sponsored by integrators or solutions providers and-especially-the unbelievable pain air travel has become in the US.
Vmworld is going to be different, though. Microsoft has already shipped Hyper-V and plans to ship Virtual Machine Manager in September, signalling that all its tubes are loaded for a showdown with the virtualization mothership.
Battles between secondary vendors are popping up here and there, and look like they'll spread. And now the tertiary vendors are not only weighing in, they're starting to elbow each other for position for what many expect will be a rush of VM-related buying and integration starting this fall.
I'm not sure it's going to be nearly the windfall most of the analysts and many of the vendors apparently do; but there's going to be a lot of effort going into the effort not only to build, but sell technology to customers looking for ways to consolidate their server farms and cut the management costs associated with them.
Interesting set of dynamics, that. Lots of opportunities for bargains and special-support contracts for customers who know what they want, and lots of spitting and trash-talking from vendors worried about being left out of what's turning out to be not just a part of a big market, but a big part of every IT market.