Archive for May, 2009

Retrospective, What is New is Old?

Looking way out there

Recently I attended my 25 year high school reunion. It was not an earth moving experience, but enjoyable to my surprise. Many of the friends I saw at our 10 and 20 year reunions I saw again. Although I was a bit of a geek, and an outsider to an extent, definitely not part of the in-crowd, but I was not an outcast either. I was just normal enough to be accepted by the brains and jocks. I realized returning home, later the same evening of the reunion dinner, that I was successful, that I have done well. Sure, there are always some individuals that do better and some that have done worse in life, and I have never been one to keep score, I did realize, that enough time has gone by, for the next generation to be learning how to find their own way, their own path. That we are no longer the next generation, but the past generation, it was ironic, that given that I work with the cutting edge of computing, that I am still part of the next generation, in a computing sense, if not in a chronological sense.

Of course, my thoughts turned from my personal career and the information technology industry in general. I consider myself as a strategic thinker, wanting to believe that only the best years for the industry are ahead? That robotics and man-machine-interfacing will continue. The science fiction fan in my, wants to see bionics, cybernetics, nano machines, etc. all benefit us. But I cannot in good faith, wear my rose colored glasses, any more when it comes to virtualization. Taking off said glasses allows me to look at things with an honest perspective, and what I am about to say will not make some people like me, or my views, fine, I am not here to be loved, liked, or even respected, if I am not honest. Of course this is my opinion, of course it is just my perspective, but in the past, my perspective has been accurate based on what the facts of history are or became over time. I leave the readers of this blog to call me to task, or yell back as they desire, saying I am insane or crazy, but so be it. Here goes. Let history judge:

  • Xen will not survive, not because it is not good, it is, but because market share trumps all. Citrix has not owned the low cost virtualization market segment, and now with other competition maturing they are under severe threat. Xen also is not free and compared to KVM and Hyper-V is not that cheap, at least not as cheap as one would wish.
  • Hyper-V will struggle for another two years, because it is still weak. Being free only gets you so far. Microsoft will make Hyper-V successful, and as with how Microsoft all but destroyed Novell in the distributed server market, many managers in information technology do not have the balls, yes lack the balls, to not pick Microsoft.
  • Life-Cycle wars, is now, and only now gaining a beachhead in the mindset of management. This is not because it has been a concept missed or ignored. But because the solutions that implement life-cycle management have been horrible. They have lacked insight and flexibility. To an extent, until some given firms have 1000s hosts, or say 5000 virtual instances, management is a nice to have, not a key requirement for many, because focusing on growing the virtualization environment, is not the same priority as managing the environment?
  • Entrenchment is the ruling strategy, or retraction might be a better term. Collapse and consolidation are the only strategic plans that anyone in the corporate world seem to believe it. The current economic situation worldwide is just an excuse to accelerate what was already going to happen. Virtualization promotes consolidation, so now that firms can compress computing infrastructure just as they can reduce man power? In fact, virtualization has slowed cloud computing not encouraged it.
  • KVM is going to grow and strengthen, but will only do so, as long as it is dirt cheap and, at the very same time, matures to incorporate key features that most end-users of virtualization must have, including, basic virtual instance life-cycle management, self-serve virtual instance provisioning, ease of use, transparent live migration, and some type of master image, common boot image, cloning and template functionality, and of course at some point a very lean, near embedded hypervisor as well as a strong virtual container model.
  • Of course, last but not least, VMware. Well, this is actually obvious, at least to me it is. I made the point to a friend at VMware over two years ago, when Hyper-V was nothing but a bad dream on a drawing board at Microsoft. VMware just costs way to much. This view of mine was reinforced in a recent meeting with VMware, where the discussion of VMware feature set, and the associated pricing became, well, to be fair, enthusiastic to be sure. It was professional, it was honest, and it was quite clear, that VMware was not hearing us. VMware has for the last 5 or 6 years, continued to add features, failed to enhance existing features in reference to scale and scope, for enterprise clients. VMware still thinks and acts like the small customer is their strategic niche. VMware has not, in my view, learned anything from its key enterprise customers, when it comes to pricing. To be blunt, VMware increasing its pricing, now, in this economic situation, is not good, and is not going to work. Not based on the feedback I am getting. When a cheaper solution has 60 or 80 of the VMware feature set, but VMware costs even more than ever before?
  • Last, we have not seen the next great quantum leap. The next great wow is still out there. Microprocessors have hit a wall; we are scaling in the horizontal faster than the vertical. Memory and storage density need a break through, we still struggle with the basic laws of physics, but sooner or later, chemical or biological computing will emerge, maybe as part of a break-through in man-machine interface design effort. But virtualization on cell phones, as cool as that is, is not good enough, just a re-packaging of what we already have.

What is new is old, why did I say this? The same basic issues for information technology still exist, getting more for less, getting higher performance for less resource commitment. In fact, the same key issues for virtualization still exist. Virtualization is still complex, yes, complex. Management of virtualization environments suffer from complexity and scaling issues. This has never changed. As the demand for larger and faster virtual instances has grown, the monitoring and maintenance of these same environments is not easier, it is harder, just the scale and scope of growth alone have far out stepped any rational expectations. In fact, the information technology industry has done this over and over. With each new reinvention of the same original concepts, gains critical mass, it happens. The move from mainframes to microcomputers, from centralized computing to distributed server computing, from the distributed server computing refactoring back in to centralized processing or consolidation, yes via virtualization to achieve cost avoidance, we create our own challenges.

This swing back and forth is normal, it cannot be avoided, we improve performance, only to realize we spending, and we reduce expense only to realize we now must have improved performance, on and on. But the days of purchasing the best of breed solution are gone, just because the best of breed is just that, is no longer defendable, welcome to a key aspect of retraction! We are going to have to settle for the Just-Good-Enough (JGE) mindset, and that puts severe pressure on VMware, and will force minor entities out of the market, so KVM, Xen, Parallels, Hyper-V, etc., will all fight for market share, below the VMware price model, offering a smaller feature set, but under cutting VMware with vengeance.

My 92 year old Grandmother is fond of saying… There is nothing new under the sun. This is either the result of 92 years of life or that she is a quite a biblical scholar, maybe both. But it is true, I hate to say it, but right now, at least in reference to strategies around virtualization, that is true.

3 comments May 14th, 2009


May 2009
« Apr   Jun »

Posts by Month

Posts by Category