VMware: CEO Greene out; Revenue light

VMware: CEO Greene out; Revenue light

Summary: VMware on Tuesday swapped CEOs and said that its revenue growth would fall short of expectations.In a statement, VMware said that CEO Diane Greene has departed.

SHARE:

VMware on Tuesday swapped CEOs and said that its revenue growth would fall short of expectations.

maritz.pngIn a statement, VMware said that CEO Diane Greene has departed. She will be replaced by Paul Maritz (right) effective immediately. Maritz will also be president and sit on VMware's board. Maritz had been president of EMC's cloud infrastructure and services division.  Maritz retired from Microsoft in 2000 and had managed the development and marketing of the software giant's key products, including Windows 95 and Windows NT.

As for the revenue outlook, VMware said that it expects revenue for 2008 to "be modestly below the previous guidance of 50 percent growth over 2007." Wall Street was expecting revenue growth of about 51 percent to about $2 billion for 2008. For the second quarter ending June 30, VMware is expected to report earnings of 23 cents a share, according to Thomson Financial. For 2008, VMware is expected to report earnings of $1.04 a share.

Since going public, VMware has had a lumpy quarter, but for the most part has been a relative technology darling as the march to virtualize technology infrastructure continues. It's unclear why Greene is departing another quarter of weaker than expected revenue growth couldn't please VMware's parent EMC. Joe Tucci, chairman of VMware and EMC, said that Maritz's has the track record "to lead VMware to its next stage of growth and development." The subtext: Greene couldn't get VMware there.

The timing of Greene's departure is notable. Microsoft is rolling out its Hyper-V effort, which isn't as good as what VMware offers, but good enough to cause some serious disruption to the virtualization market. Meanwhile, VMware has been sealing long-term deals with customers to protect its turf. Given the impending threat to VMware perhaps it only makes sense that the virtualization software company would recruit a Microsoft alum to move it forward.

In any case, the market didn't take the news very well--knocking VMware down about 30 percent.

vmw.png

Topics: Legal, Banking, Enterprise Software, Operating Systems, Software, VMware, Windows

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.

Talkback

19 comments
Log in or register to join the discussion
  • VMwhere

    Virtualization is a crock. It makes perfect sense to single-tasker people that just can't understand how more than one application can run on a server. But I digress - and I won't elaborate on this.

    VMwhere is great for Development environments. You can whip one up, play with it and get rid of it. This saves money as "real" servers don't have to be subjected to these gyrations. It is also a nice medium for setting up your entire application + OS and then offering it to others to take a look at.

    But all of this stuff can be done with the FREE versions of VMwhere. I use the VMwhere server at home and it works just fine for me. I use it to run Windoze apps (games) on my Linux desktop. VMwhere works well as a SoftWindows/SoftPC-like emulator (of the future!).

    Using VMwhere ESX server in a production role makes little sense to me. I was able to bring the thing down by "exploiting" its weakness - emulated networking. Try having a couple thousand SSH sessions try to connect to a VMwhere VM and you'll see what I mean.

    It's a great little tool for what I outlined above, but I wouldn't want to run my enterprise on it.
    Roger Ramjet
    • virtualization is most certainly not a crock

      I am running 15 servers on 4 boxes using 3tb of san. Using vmware server. I would use ESX, except it is kind of expensive and ugly for the windoze noobs who may have to click stuff for me when I'm not here.
      It makes the enterprise infinitely more manageable, in addition to the obvious hardware/power saving and system portability. I can add processors, memory, drives, etc - all from my chair that's not in the server room. the only reason I go in there is to change tapes. I could go on...
      erm2
      • Country Crock

        If you run X applications on 15 VMs on 4 boxes, why not just run X applications on 4 boxes? Just what does the overhead of VMs do for you? Other than warm fuzzies, I don't see the need.
        Roger Ramjet
        • Surely not a crock

          Rule of thumb. For normal x86 based application deployment, normally ISV's will state that one application should reside under one OS and it cannot be shared with other application.

          When combining the X(A+B+C...+Z) applications into 4 servers, most noticeably is when something goes wrong with application A, the app A support team will surely point out that app B and C is giving those errors.

          Besides, using ESX, the overhead is minimal. The image is as low as 32MB(using ESX3i) thus giving more resources to the residing VM's.

          To wrap things up, 99 out of Fortune 100 (if not all 100) uses VMwhere? for production use. In this case, I agree that virtualization is not a crock but it is a gift for us to consolidate and contain our servers from sprawling!

          :)
          scalper-din
          • Crock is what Crock does

            For a human being to concentrate on doing one thing at a time - VMs make "sense". But computers don't work like that! *NIX OSes are made to multitask - which means running more than one task at a time (I thought you might need an explanation). Tasks are independent of one another - so if one goes "haywire", the others are not (directly) affected.

            The trade off becomes this: If a haywire task grabs all/most of the server resources, the other tasks are affected - so if it was isolated in one VM, only that one VM is affected. BUT...
            Multitasking has the nice ability to assign priority to tasks. Everything works very smoothly as tasks share CPUs, memory, disk, etc. Under a VM, all of those things become MANUAL PROCESSES. TO share CPUs between VMs (use IBM AIX example), you have to create "pools" of CPUs and assign rules for how each VM uses those CPUs. The same thing goes for memory. SO you end up MANUALLY chopping up resources for VMs whereas the original OS handled all of that AUTOMATICALLY.

            How much work do YOU want to do, Crock?
            Roger Ramjet
          • I hope you aren't serious...

            So what do you do when a physical box goes down? Oh, you forgot about that tradeoff I assume! When a VM host goes down, the virtual servers are moved automatically among the other hosts.

            [quote] you have to create "pools" of CPUs and assign rules for how each VM uses those CPUs. The same thing goes for memory. [/quote]

            So what? How is this different than building/buying a physical box that meets the hardware requirements for the software you are installing on it? Are you saying you just wing it? Buy the biggest box and throw as many CPUS / Memory as possible and see how it works?

            VM allows to you micromanage resources so you can get the most out of them.

            If you manage servers. (Which I hope you don't) I'd love to see the utilization/performance logs. I can only imagine how much power you are wasting.

            You are right.... all the fortune 500 companies that virtualize are absolutely crazy, maybe they should hire you to set them straight.
            NetworkBankAdmin
          • Crazy is as Crazy does

            [So what do you do when a physical box goes down? ]

            All the VMs that reside on that box go down.

            [How is this different than building/buying a physical box that meets the hardware requirements for the software you are installing on it? Are you saying you just wing it? Buy the biggest box and throw as many CPUS / Memory as possible and see how it works?]

            Buy the biggest box and keep adding applications until you get the best utilization. Then buy box 2. You can do the "math" ahead of time to see exactly which apps to put on a box - but how does the plan and implementation line up afterward? The "only" difficult part is how to charge the departments for the hardware . . .

            [VM allows to you micromanage resources so you can get the most out of them.]

            I have yet to see any area where micromanagement is the best strategy. Why micromanage when the OS has everything built in?

            [If you manage servers. (Which I hope you don't) I'd love to see the utilization/performance logs. I can only imagine how much power you are wasting.]

            If you run 10 VMs with 10 apps on a single box OR you just run the 10 apps on that box, what is the difference in power utilization?

            [You are right.... all the fortune 500 companies that virtualize are absolutely crazy, maybe they should hire you to set them straight.]

            I'm just doing the (simple) math here. There is absolutely NO advantage (in utilization, power, cost, etc) to running VMs on a box if you use the same number of apps (for a non-VM box). You ARE adding complexity and cost.
            Roger Ramjet
          • K-Rocks!

            Yea. Imagine, *NIX(in this case linux) under those VM, even more tasks can be run, optimizing the full potential of the x86 processor(s) capabilities and not to forget the memory(s) sitting there idling with utilization around less than 10% even with your task(s) running at peak.

            The real haywire is when you tell the application vendor/principal support that you are running their solution along with some other solution as well. You will end up liasing with all the related softwares runnning in it. If it is open source, then more mailing lists, more forums and more emails for you to go in order to determine the real wire!

            For actual production uptime issues, this is where the VM HA and VMotion kicks in, when one VM goes wrong, the standby node kicks in AUTOmatically, and oh yea, need some manual configuration initially.

            In dynamic resource mgmt, VM DRS is the key. When you need more resources, the VM can be moved to a new free ESX host, automatically. But again, it needs manual config via some wizards to enable this.

            For AIX, in this case, that is the beauty of PowerVM. With POWER6, you can even do micropartitioning of your CPU up to 0.1 of it. And yeah, the LPARs need manual config and system study is needed to determine the initial allocation. You surely dont want to run only one instant in a System P box.. except you are runnning a VERY ENORMOUS DB of somekind.

            To conclude, workloads will grow exponentially. Believe me, virtualization is the Key to less ROCKS being arrange for a new DC expansion! Less power, less cooling, less space, less hardwares, simplify mgmt, and the list will go on.....

            Just my 2 cents. :)
            scalper-din
          • Lies and illusions

            Do you understand fragmentation? In terms of disk fragmentation, you have a single file that is broken up in pieces - and each piece is "scattered" over the disk. DE-fragmentation is taking all of those pieces and lining them up next to each other. This makes for more efficient disk transfers, as all the file data can be "swooped" up in one pass of the read head.

            An UNIX Operating system is multitasking. This means that many tasks run concurrently. The kernel is optimized for this purpose. These tasks run with a certain priority - thus you can give one task priority over another. Also, the way multiple CPUs work is that tasks grab available CPUs automatically (also memory is allocated automatically).

            When you create a VM, you negate all of these things. You have to MANUALLY spell out all of those items (who gets CPU, memory, priority). It can be very fine-grained like that Power6 0.1 CPU share. So you are telling me that HUMANS are MORE capable of deciding what resources to allocate - then the OS is? Sounds great for sysadmin job security, but is it the best for the company?
            Roger Ramjet
    • Mainframes, Solaris, AIX ... even Windows

      Right ... which is why AIX has LPAR's, Solaris has LDOM's and zones, mainframes have VMs, and even MS is virtualizing.

      In my Fortune 50 company, virtualization is a big deal, where every production server which can be virtualized IS being virtualized. No matter what your opinion is on virtualization, it is not aligned with reality.
      davidr69
      • Know what a "fad" is?

        I am well aware at what the industry is doing. I have personal experience working with VMwhere (and LPARS) (I have a VMwhere v1.0 license!). In non-production, development areas - it makes a lot of sense to use VMs. But the hype doesn't hold water when it comes to production servers.

        What DOES make sense is that sysadmins see this fad as job security. More moving parts = more mechanics needed. I can see why my words produce anger and insults - rather than analysis. Look at who's doing the insulting . . .
        Roger Ramjet
  • RE: VMware: CEO Greene out; Revenue light

    As a layman, my understanding of VMWare???s triumph is just opposite to today???s cloud computing. Their core concept was to simulate a given hardware into multiple computing units. But here, cloud computing is more about integrating all given hardware to one giant computing unit. What is Martiz planning to do? VMnoWhere!!! ;-p

    [Saurabh Kaushik]
    http://saurabhkaushik.wordpress.com/
    saurabhk666
  • Virtualisation and cloud computing aren't mutually exclusive

    Virtualisation is about the foundations of computing. It
    allows you to run multiple instances of an OS on one box -
    on servers it's usually multiple instances of the same OS, on
    PCs it's different OSs.

    Cloud computing is about access to data and applications,
    it's not about how the cloud is built. An analogy is that
    virtualisation is about the roads, cloud computing is about
    the transport system whose vehicles run on the roads.
    Fred Fredrickson
  • RE: VMware: CEO Greene out; Revenue light

    We currently have 47 virtual servers running on 4 VM ESX servers.

    The advantages we see...
    1) Power usage cut in half (pre-virtualization)
    2) balance the performance of our app servers
    3) central administration of ALL our servers
    4) ZERO DOWN TIME
    5) New server up in minutes, not hours.

    <sarcasm>You're absolutely right, it's pointless and a "crock" to use virtualization. </sarcasm>
    NetworkBankAdmin
    • Nice

      Totally agree with the advantages you lined up. Now, we can scale UP and not scale OUT! :D
      scalper-din
  • RE: VMware: CEO Greene out; Revenue light

    Man.. I wonder how things will go from now on. The evolution is so fast and for myself, keeping up is the hardest task!
    scalper-din
  • RE: VMware: CEO Greene out; Revenue light

    I hope it goes well as I have sunk a bit of retirement bucks into VMW!!!!!!!
    sleep well puppa
    puppadave
  • The truth.. about humans..

    Refers to the last thread by Roger Ramjet.

    Disk fragmentation is another set of thing which I am not discussing here. True, Unix variant was built to multitask. You are right. And it was built by humans.

    I wont line up any more arguments here, but one final question for you. Are you the one who did the sizing for your *optimized*, *big*, and can handle *loads of app* concurrently?

    Well, somehow, HUMAN will count into action, isn't it? Very capable, right?
    scalper-din
  • Virtualization has always been a flop

    Way back, IBM had its Virtual Machines. Virtualization only lets the user prolong the day when everything gets properly upgraded. Worse, it lets the user sink deeper into the morass of a messy bunch of virtual machines - all of which require knowledge and maintenance.
    Virtualization is a trap, which rhymes with crap.
    John Harding
    http://www.escapefromparadise.com
    jbharidng