July 9th, 2010 // Cloud Com­put­ing, Soft­ware as a Ser­vice // Justin Alexander

The state of the cloud

It’s hard to find a tech­nol­ogy mag­a­zine, blog or por­tal today that doesn’t con­tain some arti­cle or adver­tise­ment for cloud com­put­ing. It’s even on the evening news!

All of this excite­ment is for good rea­son. The migra­tion of com­put­ing capac­ity from pri­vate cor­po­rate net­works to shared pub­lic clouds rep­re­sents a huge oppor­tu­nity to make IT depart­ments more respon­sive to the busi­nesses they sup­port. New tech­nol­ogy solu­tions can be deployed more quickly, addi­tional capac­ity can be added to exist­ing sys­tems “Just in Time” (JIT), and unneeded capac­ity can be removed before the next billing cycle begins.

The abil­ity of cloud host­ing providers to meet demand for het­ero­ge­neous tech­nol­ogy solu­tions using a generic pool of com­put­ing cycles also promises to dra­mat­i­cally lower costs for com­pa­nies of all sizes. Com­put­ing capac­ity appears to be nearly unlim­ited, and cus­tomers only pay for the resources that they use.

The ana­lysts cer­tainly agree. In a recent study, Gart­ner pre­dicted that cloud com­put­ing would grow from $58.6 bil­lion in rev­enue in 2009 to $63.8 bil­lion in 2010. By 2014, the global cloud com­put­ing mar­ket is expected to be an impres­sive $148.8 bil­lion. That rep­re­sents an annual growth rate of more than 16%.

A mod­ern land grab
But the cloud is not lim­ited to sil­ver lin­ings. In fact, there’s a lot of vapor up there. This is an imma­ture mar­ket that has just com­pleted its sec­ond wave of incu­ba­tion. There are no clear lead­ers yet. Lit­er­ally thou­sands of star­tups and estab­lished ven­dors are vying for their piece of the pie. Com­monly accepted stan­dards, oper­at­ing pro­ce­dures and legal prece­dents don’t exist. We’re effec­tively wit­ness­ing a vir­tual land grab, not unlike the West­ern expan­sion expe­ri­enced in the 19th Century.

A point of con­ver­gence
It’s easy to be skep­ti­cal about such glow­ing pre­dic­tions. Indeed, no one – not even Gart­ner, can reli­ably pre­dict the future. How­ever, I per­son­ally believe that the cloud com­put­ing ser­vices trend is real. The most com­pelling evi­dence at my dis­posal is the obser­va­tion that cloud com­put­ing isn’t a sin­gle prod­uct or even a spe­cific ser­vice. Rather, it the con­ver­gence of sev­eral tech­ni­cal and busi­ness trends that, when com­bined, cre­ate the basis for a new gen­er­a­tion of com­put­ing solu­tions that solve real-world prob­lems. In no par­tic­u­lar order, these trends are:

Vir­tu­al­iza­tion
Sim­ply put, the cloud would not exist with­out vir­tu­al­iza­tion because it allows the cloud to scale effi­ciently. Vir­tu­al­iza­tion is what trans­forms a server from a single-tenant plat­form ded­i­cated to a sin­gle pur­pose into a generic set of com­puter cycles that can be con­sumed by any­one to accom­plish anything.

Gart­ner pre­dicts that 60 per­cent of server work­loads will be vir­tu­al­ized with an aver­age den­sity of 10 servers per phys­i­cal server by 2013. That equates to 5,708 new vir­tual machines being cre­ated each and every hour of every day. They also pre­dict that this work­load will be achieved using only 10 per­cent of the total num­ber of phys­i­cal servers sold within the same time­frame. In other words, 60 per­cent of the world’s com­pu­ta­tions will be accom­plished using only 10% of the phys­i­cal server capacity.

In my next blog post, I’ll share more trends that are lead­ing to new com­put­ing solutions.

Read more:http://www.linkedin.com/news?viewArticle=&articleID=148422657&gid=37569&type=member&item=24424302&articleURL=http://blog.hyland.com/saas/part-i-%25e2%2580%2593-red-herring-can-fly-hyperbole-in-the-cloud-welcome-to-the-wild-wild-west/&urlhash=H71w

Com­pli­ments of File­Man Research