VMware is making a play to wrest data centre control away from every other vendor and become the single door through which users and admin staff see data centre server, networking, and storage resources. A data centre administrator will increasingly monitor, manage, and diagnose data centre resource problems through VMware APIs …
As I am currently working on a massive commercial implementation of Citrix, and being somewhat aware of their market penetration and love for their Xen hypervisor I would have to disagree with your conclusions. I do think that you have the general sentiment correct, that Hypervisors are becoming the new centralization point for the Data Center, but I think you paint a partial picture.
Sun Virtual-Box, MS Hyper-V and of course Xen and others will not just take this lying down as you seem to imply. These companies have already seen, or will see the writing on the wall.
Does VM have a leg up over the competition? Sure it does now, but I think it is naive to think that all will necessarily stay that way.
VMware has a great collection of tools allow you to centralize your management. They flaws with you comparing them to M$, and assuming a coming VMware dominance is:
1.) The key word is ALLOW, with M$ for the most part if you wanted your documents universally able to be read you NEEDED to have M$ office. This monopoly factor that helped M$ so much does not exist for VMware. If they raise prices, or just by nature of being dominant people will migrate to other options or create them. People just don't NEED to have VMware.
2.) Their product is not unique: Citirx and Sun Thin clients also can virtualize almost ANY OS or application out there .
3.) Storage vendors will either be able to continue making money of their proprietary solutions, or if forced will move to adopting more general storage open standards. If this is the case, Xen, Virtual Box and others will have the same available access to these API connections that VMware does.
4.) As some of these programs like Xen are open source, I would argue that companies like Dell, IBM, Redhat, are likely to adopt the code and modify it for their own releases.
5.)Your argument ignores more than 1/2 of the Data Center-Enterprise market.
VMware and virtualization mainly deal with two major sections, Servers (applications and DB) and Storage. The network is typically a completely different set of management tools. In most corporations I have worked with, these are sectioned into entire different departments of people, making interoperability less important. CiscoWorks, Nagio, Sitescope, HP OpenView, and others have been working for decades to create a single interface in which to just manage your networking elements. Further there are a ton of new devices (Packet Shapers, Load Balancers etc from compaines like SilverPeak and BigIP) that don't as easily fit into these other companies management systems. BlueCat runs the best IPAM (IP address management) software currently and I don't believe you can manage the boxes from anything other than their management console. And then there is environmental and other monitoring. Not to mention all the current, and future security devices.
5.) MIBS - Most servers, storage and networking devices do have universally available MIBS for SNMP reporting - I really see this is a much more ubiquitous tool in the DataCenter, and like Ethernet, TCP/IP and other widely used protocols, it is not owned by any one company.
You mention the exceptions in your article but don't really spell out the celestial bodies that are already in VMware's orbit. Which companies do you consider to have already drunk the Kool-Aid so to speak?
Oh and as I understand it, Virtual box is capable of using VMware machine images, if VM machines remain on a standard open format, not only is security a issue, but it will also be extremely easy for companies to migrate their entire operations from one Hypervisor to another competing one (which is how a free market should really work anyways :-)
Be EXTREMELY careful as a journalist of trying to predict the future. There is an article every week claiming that someone has cornered the market, or that X is now dead. Almost without fail they always turn out to not quite work out that way. Despite the old adage, people do still make buggy whips (even if they are serving a different market than horses). The one thing that has been proven time and time again, is that people will innovate and they will continue to compete. And no matter how great or dominant a company gets, eventually it gets fat and complacent.
Hey, but other than that being wrong, great article! No really, I do think you had some good points and shed light on a certainly interesting development in the general Data Center world.
keep it up!
Hardware is getting cheaper and cheaper. You can buy boxes to run Xen or Hyper-V for the price of the VMWare licenses alone.
You can buy Quad boxes with 128GB of ram for about the same price as the VMware license on a quad box.
It's Xen not XEN
Why do people insist on writing XEN instead of Xen. If you're in any doubt about the correct way to write it, look at xen.org, where it is quite clear that it should be Xen...
Oracle... better chance at world domination...
Victim of their own success?
Personally I think the Hypervisor is the new OS analogy and likening VMware's growth to that of DOS -> Windows in the 80's and 90's is a little flawed.
The "market share creates more market share" feedback loop worked for Windows because developers had to make a choice about which platform they were going to develop for. Once MS had a certain portion of the market, if you were a software house looking for the best returns on your product development costs, it made sense to focus on OS with the majority market share rather than burden yourself with the expense of developing for several different platforms.
On the other hand, the hypervisors greatest trick is convincing the underlying workload that it's not there (not unlike the Devil I guess ;) ). By function of design therefore, any Hypervisor is almost invisible to the virtualised OS, applications that run on said OS and the end user. This invisibility makes the hypervisor relatively easy to replace with one that is better/cheaper/faster/hardware embedded. Management and end users don't care about the virtualisation platform their applications are hosted on. All they care about is that they can collect their email, run their CRM system and browse their intranet. If today these workloads are hosted on a VMware hypervisor, fine. If tomorrow it's hyper-v or some type of hardware integrated virtualisation, so long as everything works, who cares.
VMWare almost single-handedly created the x86 virtualisation industry, and took it from a niche market to the massive industry that it is today.
The only problem for VMware is that in doing so, they have created an awareness of the benefits of virtualisation and a demand for products that can satisfy that demand but that product does not specifically have to be ESX / vSphere. Thanks to the invisible nature of the hypervisor, anyone can have a crack at building a competing virtualisation stack, and if its good enough people will use it. This is not like having to migrate from Windows to OS X. You can just pickup your workloads and drop them on to a new stack. So long as it's reliable, Exchange will keep Exchanging, and users will keep receiving their email.
VMware have already done the hard work of making the market receptive to the concept of virtualisation, now anyone who's competent can build an offering and benefit from the wide industry acceptence x86 virtualisation now enjoys. Thus, VMware risks becoming a victim of its own success.
If VMware wants to stay relevant into the future, they need to do what every good software company does, and find tactical ways of adding proprietary lock-in to their product. And that's a challenge that I think they remain to successfully address.
El Reg ...... where all the Big Hitters hang out/tune in and turn on ?
"When all servers are X86 servers then the supplier of the main server control software, the hypervisor, calls the shots."
Oh? I don't thinks so. The smartest user and/or smarter users of main server controller software call the shots/create the plays/program the Virtual Realities, actually.
"Be EXTREMELY careful as a journalist of trying to predict the future." ... By Rob Dobs Posted Tuesday 1st September 2009 20:23 GMT.
Hmmm? Sound advice much better directed at Naked Short Sellers and Financial Whizz Kids for journalists to report on, Rob Dobs, whenever Everything Revolves around Boom Busting Markets and Bust Boom Marketeers.
And that was a nice deep post, Colgate, Posted Wednesday 2nd September 2009 03:43 GMT, which I wouldn't disagree with at all. And it is definitely Master of the Universe/Holy Grail stuff.
Ovirt Dot Org
That's an extremely perceptive article - if VMWare abstraction essentially becomes the virtual data centre then that will answer huge number of data centre management issues. I'm not quite so convinced about the hyperviser usurping the role of the operating system by providing an applications with direct resources. However, it is very easy to imagine databases, J2EE environments and the like delivered as virtual appliances with an integrated highly tailored and optimised operating system layer designed to run directly within a VM. At the moment this is a tricky thing to do with "real" x86 hardware due to the install process needing to tailor itself to a massive range of different device drivers and other hardware environmental factors. By running on a VM and going through the better-controlled VM abstraction layer that problem is vastly reduced.
As far as supporting non x86 processors goes, there is an answer to this, at least for the support of legacy applications not requiring efficiency, and that is hardware emulation/run-time code conversion in the hyperviser. It's a tried and tested method that has been used at the application layer for migrations between processor architectures. It has also been used for emulation right down to support for operating systems - in the past, UNIX vendors have provided mainframe emulators (indeed there is a very good freeware mainframe emulator). Porting applications is the ideal way round this, but it isn't always practical or cost-effective. As the performance of x86 CPUs becomes ever faster, emulation approaches that were previously considered unviable become practicable.
It's not going to be suitable for high-throughput systems, but every large shop is going to have its legacy of old hardware with processors which are either out of development or whose future looks dubious. That includes, of course such processors as VAX, Alpha, PA-RISC, MIPS (any of those left in production?) and arguably SPARC or Itanium. Many shops also have a rump of IBM mainframe or Power machines which they would rather eliminate as they've moved strategic applications to Linux or Windows. Of course the bugbear here is the rights holders for the operating systems and other licensed software as costs, support and legal obstacles become an issue.
The most promising candidate for such emulations was snapped up by IBM in the form of Transitive. I rather suspect that an emulator of some sort embedded within VMWare would be considered to be highly desirable by many shops, and I wonder if this is being looked at (clearly Intel could do things to assist emulation in hypervisers as well). Emulation is just the ultimate level of virtualisation.
No monopoly on x86
VMware isn't going to have a strangehold.
Virtual machines are pretty easy to move between hypervisors, because they're all just emulating a vanilla x86 machine. So if VMware licensing gets excessively expensive, then people will first start moving their test and development environments to standalone (free) ESXi instances; then to other virtualisation platforms like KVM or Xen; and the ultimate sanction is to move their live platforms across too. The other vendors have a good incentive to provide tools to make this as seamless as possible. It's not like moving an app from Windows to Linux, or even from one Unix flavour to another.
"There is the prospect - there surely must be the prospect - of apps being produced which request their previously Windows-delivered resources direct from ESX. Every step in an app's resource consumption stack needs physical host server cycles. Why not minimise the number of steps and have apps run more and more in VMs that have a thin or almost non-existent O/S layer between the app and the VM?"
There appears to be a misunderstanding here between the services provided by the real or virtual hardware (e.g. block devices) and the services provided by the OS (e.g. filesystem). They are complementary.
Of course, vendors can and do distribute applications as .vmdk images. That gives a minor advantage to VMware users, but other platforms can import and run these images too, because they are little more than blobs of disk space.
These images are almost certain to include some sort of OS, because VMware on its own isn't going to provide OS services such as filesystem, scheduler, virtual memory etc. But regardless of what sort of OS it bundles, the .vmdk image would still be portable.
The only "OS-less" x86 app I can think of is Netware, and that's really an OS in its own right.
"Microsoft provided a good basic and cheap operating system for IBM PCs"
Good: Technically NO - compare IBM's PC-DOS which was supposed to be a rebadged MS-DOS but usually worked a lot better due to IBM applying its own standards
Ethically NO - It was imposed on hardware suppliers by M$'s infamous obligatory licensing deals which gave end-users the choice between using a bad, already-paid-for-and-installed OS used by umpteen gazillion other dupes who couldn't possibly be mistaken, or shelling out a few bucks for someone else's OS that Worked Much Better. If you wanted to sell MS-DOS with your branded PC, you had to sell it with ALL your branded PCs, whether the customer wanted MS-DOS or not (and many didn't but fell into the trap nevertheless).
Basic: More like primitive, compared to even the 8-bit standard, CP/M. Remember M$ bought and rehashed QDOS (Quick & Dirty Operating System) to make, voila, MS-DOS.
Cheap: NO. Digital Research's DR-DOS was more then 50% cheaper and worked 5 times better.
Similar comments apply to Windows. Anyone want to join in a class-action suit against M$ for retarding the PC revolution by 5-10 years? And that is unlike VMware, read Candler.