As virtualized environments become more common, new businesses to the market are being given a great deal of information by providers and users; and not all of it is necessarily accurate. While documented evidence exists of what is known as “virtual sprawl” — the tendency for businesses to mismanage their virtual data storage — less hard data is available to demonstrate the existence of “virtual stall”. This concept refers to the idea that as virtual environments become too large, a business will find their utility “stalled”, and more VPS purchases a waste of time and money. A general rule of thumb for the theory of virtual stall has been created; once 30 percent of a company’s servers are virtual, they will begin to experience stall.
The trouble with stall is that no replicated data exists yet to demonstrably prove its existence. A number of experts in the field are crying out that it will be a large and aggressive problem in the next few years, but oddly enough, these experts are virtual server management companies, many of whom argue that only their particular brand of management can save a company from the lurking evil of hidden virtual stall.
Arguments rage on both sides of the fence — naysayers claim that virtual is just a tool created to get companies to buy managed IT products, while proponents argue that businesses must be prepared. Simple facts have shown that companies do need to be mindful of how their storage is handled and where their data is placed, but virtual stall has not yet been declared the overall villain in this piece. Regardless of the why, however, businesses need to ensure their virtual storage is at least as effective as their physical one.
Related News:
{ 1 comment }
Hi, I am really happy you have taken up this topic. I wrote a lot of the early analysis that set the concept of virtual stall (or ‘VM stall’) into the wild, so it is interesting to see it pop up in so many places, with such a wide variety of views.
However, I do take issue with your premise that there is no data to prove the existence of virtual stall, and that it is only server management vendors (like my company, CA Technologies) waning of this issue. A cursory search will reveal lot of commentary, research, and more – from vendors, yes; but also from administrators and practitioners, independent journalists, bloggers, consultants, and industry analysts (including Gartner Group, Enterprise Strategy Group, Gabriel Consulting Group, IDC, Enterprise Management Associates, and 451 Group).
Also, as I mentioned to Arthur Cole (who interestingly posted a very similar statement in his post at IT Business Edge), I for one believe it is not solvable by “only [my] particular brand of management”. Solutions like the ones from CA Technologies are important, but virtual stall must be addressed as a people, process, *and* technology issue.
If you need some pointers to supporting research, please let me know. You can start by reading the several posts and links on the topic at my blog, http://pleasediscuss.com/andimann/, or just drop me a line.
Andi Mann
CA Technologies.