Part 2 of "Some thoughts around the latest buzz of Cloud Computing"

First, it is really obvious that there is no clear definition of "cloud computing", but that doesn't seem to dampen the excitement and enthusiasm for this new computing model.

First, it is really obvious that there is no clear definition of "cloud computing", but that doesn't seem to dampen the excitement and enthusiasm for this new computing model.  I guess so that we can put some more structure around the discussion lets just say that "cloud computing" is a generic term for any information technology solution that does not use in-house data center or traditional managed hosting resources.  Sound fair?

So here are some common threads and key differentiators as I have seen them:

    * the concept of providing easily accessible computing and storage resources on a "pay-as-you-go", "on-demand" basis, from a virtually infinite infrastructure managed by someone else.
    * The capability to access your application anywhere, move it freely and easily, and inexpensively add resources for instant scalability.
    * In my opinion it represents a true next phase of computing, and it is changing the way IT infrastructure is being delivered and consumed.

So back in the day before I was in IT, I lived another life as a trader on Wall Street.  I still have some friends out there so I decided to make some calls to see who could hook me up with some of the analysts that track this industry.  Well it turns out that this market is hotter than I originally had thought.  Some of the analysts put cloud computing at more than a $160 billion dollar market.  Holy %&*^!!!  Really? I know that's a pretty big number isn't it?  But get this, even though this computing model is still in its infancy some of the leading indicators are really showing how fast this model is growing.  For instance, Forrester reports that bandwidth for Amazon’s EC2 and S3 in Q4 of 2007 exceeded all of the global Amazon.com web properties combined during their busiest time of the year.  Again, holy $#*&%$!!!!

Ok, so now I'm wondering what are the types of products that are available in the cloud.  I use the word "products" semi-loosely.  So I did a little more research and found three types:

   1. Applications in the cloud (Salesforce comes to mind first)
   2. Platforms in the cloud (Google's AppEngine comes to mind first)
   3. Infrastructure in the cloud (Amazon's EC2 infrastructure comes to mind first here)

There was a conference just last week (June 25th) called Structure 08 in San Francisco.  I wasn't there this year, but you can count on seeing me there next year.  Anyway, one of the opening addresses was by Jonathan Yarmis who is the VP of Advanced, Emerging and Disruptive Technologies at AMR Research.  Mr. Yarmis echos what I have been blogging about for a while now, Change. Mr Yarmis states that, “The world is about to change, and change in profoundly interesting ways.” What have I been saying?  He goes on to state that, “The enterprise itself hasn’t figured out how to embrace cloud computing; users are figuring it out very quickly.” How true is that?  If you think about it users are way ahead of where the rest of enterprise IT is and IT needs to be paying closer attention.

I'm really starting to dig in on this and getting my head around what is going on in this space and what this will be doing to our world (Citrix, VMware, etc etc).  I mentioned it in another post, Virtual Service-Oriented Grids.  I can guarantee as sure as the sun will rise tomorrow that within the next five years we (being IT professionals) will be entering in or already neck deep in a transformational period due to the new architectures of enterprise IT.  It might be time to bring your skills up my friends.

Join the conversation

9 comments

Send me notifications when other members comment.

Please create a username to comment.

good stuff; Hosted Services are the way of the future; death to traditional IT!
Cancel

Hi Michael,


I must admit I have been reading more than a few articles around this very subject in the last few episodes of Wired - I've just tried going back and find the appropriate articles/links but have not been very successful.


It is a very seductive idea and I only wish I could reference some of the details now.....

Cancel

I have had an interesting book cross my path in the last few weeks, http://www.amazon.com/Microsoft-2-0-Plans-Relevant-Post-Gates/dp/0470191384/ref=pd_bbs_sr_1?ie=UTF8&s=books&qid=1215013931&sr=1-1


The dialog regarding cloud computing references Microsofts push into the MSP market.  The initiative behind Microsoft's Live Services is all about the cloud.  According to the author, the Ever Ready battery company (the one with the bunny beating the drum on U.S. television), has become the first major industry to move all of its network operations into the Microsoft cloud environment.


I don't know about you, but I have had hotmail accounts for years.  I can recall only a few incidents where access was not immediately available.  Microsofts ability to provision Live Services for the consumer is driving innovation for business.  The competition for the clouds is of course Google.  I am anticipating management software which will configure access to Live services based on domain identities maintained by individual businesses.  You logon with your domain identity, and MS Live Services are immediately available to the end user is what I see.


The MSP space is HOT.  Desktone, FusionStorm, and others are clamoring for solutions to provision VDI platforms in a dynamic way for as vehicles for service offerings to end users via the cloud.  I have a bias toward such a solution...:-), which of course is RES PowerFuse, but this is another story...


The clouds are coming, no doubt about it.  Thanks for the post.


RTE

Cancel

You raise an interesting point.  In all the conversations that I've had over the past months that was a question I had to ask.  One of my colleagues threw out this unscientific answer to a very unscientific poll he took at Interop last year. He asked the audience in a session how many of them would trust Microsoft to run a cloud with Microsoft applications and how many would like to see Amazon run a Microsoft kernel in their EC2.  He told me that 75% of the audience responded that they would trust Amazon to run Microsoft's apps rather than Microsoft.


Interesting indeed even if it isn't very scientific. 

Cancel

"Change" is such a tired expression.  Whether it's politics or
technology, people try to make you think that we're always on the brink
of some radical new changes...as if we'll wake up one day and things
will be completely different from the way we knew them.  Like the flip
of a switch...  It usually doesn't work that way.  Though we might not
be privy to all the developments leading up to a new technology, such
technological changes typically have been long in the works.  Maybe
they weren't economically viable or mature enough to be released when
first originated, but they weren't created in a day. 

To me,
one of the key phrases in your article is that "there is no clear
definition of 'cloud computing'."  No one can describe in concrete
terms exactly what it is and *how* it will change things.  Even the
name "cloud computing" is nebulous.

The term cloud computing
reminds me of SETI@home...remember that?  The idea of a distributed
computing grid has been around for years.  Can anyone point to any
radical changes as a result of these efforts?

As for the alleged
impact on IT jobs, who exactly is going to manage this cloud?  Sure,
some parts of computing might become more like a service.  However,
whether the computing infrastructure is in-house or in the cloud, it's
still will require management.  So, I don't see how it's going to make
IT jobs disappear.  

And what about ownership and
accountability?  In an era of increasing regulation and scrutiny, who's
going to own the information and ensure proper security?  Will IT
organizations in heavily regulated industries really want their
sensitive data in the cloud?

Lastly, I'm not saying a
distributed computing model won't be beneficial.  But I think the
benefit will be in allowing IT organizations to focus on more elevated
work that grows the business vs. maintaining the status quo.  Because,
at the end of the day, who's going to know more about how to grow and
adapt IT to support business change: a third-party service provider in
the cloud or someone who works for and knows the business?  Like with
most things, there will be an opportunity threshold: the cloud will
work for some things, but with others, "traditional" IT or some other
model will make the most sense.

--BXP

Cancel

Thanks for the post.  You are exactly right, the idea of grid has been around a while (SETI@home, I do remember that) in many forms.  I think in today's environment it's being revisited with more attention.  "Every that is old is new again" remember that saying...

The term cloud computing is very nebulous and amorphic.  That's why I have tried to put some sort of wrapper to it so that the conversation we are having here has some basis.  

I don't think this will be an alleged impact on IT jobs, it's going to be very real.  The thing here is that these jobs will move around to different areas of the company and IT in general.  I'm not saying they are going away and anyone who says they will go away is a few doughnuts short of being a cop.  They just take a different form.

Ownership and accountability, easy.  There are SLAs, OLAs, etc in place.  It is ultimately your data and applications so I think that is ultimately back on the business.  

Security....great question.  That is the biggest ? if you ask me.  I'll take a quote from my good buddy Shawn Bass; "it's not for everyone".  There are some use cases where it may make sense to create your own cloud/grid so that this stays internal to your corporate firewall.  

Thanks again for the post, I love having these conversations.

Cheers man

Cancel

My take on this is that the analysts are doing what they get paid to do.  So lets ignore their numbers but listen to the message (with appropriate filters on: "Shields up, Mr. Chekov:).


I look at this with an eye to the ASP/MSP market that started many years ago.  The concept of taking software delivered as a service by others is not new to the name "Cloud Computing".  There are huge barriers to overcome, both technical and personal.


On the technical front, we keep seeing advances in technology that make the implementation easier.  Terminal Servers and Citrix were a tremendous advancements that made ASPs possible.  Combined with Application Virtualization it became more affordable.  But this did not drive the masses to accept the concepts.  Ultimately, if the Enterprise could afford IT staff, they basically could not get over the personal objection to having their data outside of their control. New advances on the technical side continue to help the financial aspects, as well as expand on the classes of problems that can be solved.  In particular, "Cloud Computing" may extend reach and supposedly safety/security.


Analyst and Trade Magazine fanfare will undoubtedly get early adopters to try out new offerings.  Some things will be highly successful (Salesforce.com has done so for a long while), and some will not. 


I see change happening out there, but I see it with a small number of fast fliers that make great news.  Over time, those success stories can help build support to overcome the personal objections that CEOs have.  But one bad story can easily undo a lot of good work.  So let's get excited by all this, but just keep things in perspective.

Cancel

Tim..thanks for the response.  Perspective is always good.  In the next post I made "As the Clouds grow", I point out some of the challenges that are going to be faced with this reincarnation of an old concept.

Thanks for ringing in.

Cancel

No problem!  I enjoy these convos as well. :-) 

One of the primary drawbacks to any "how x-y-z technology will change the world in x number of years" prediction is its inability to factor in other technological breakthroughs (this goes for things outside technology as well).  If you apply the grid model to a stagnant computing landscape, sure, you can foresee drastic changes.  But this assumes everything else stays the same. 

While the grid might be a solution for current challenges, new challenges will arise, which the grid might not be able to address.  Furthermore, grid alternatives (beyond "traditional IT") may emerge.  For example, let's assume an IT shop uses the grid for VDI for desktop provisioning.  OK, so now desktop provisioning and management, which used to be a function of the IT shop, has moved to the grid (and that corresponding IT job has moved to the grid).  This frees up the IT shop to focus on other things to improve the business such as improving security, compliance or flexibility.  That doesn't just move jobs around to different areas: it creates new jobs/roles.

As a parallel example, take the introduction of automatic robots into the automobile manufacturing industry.  The use of robots did eliminate some jobs, but didn't eliminate the need for autoworkers; it just changed job functions for which certain autoworkers were needed.  So, instead of needing a worker to weld part A and part B together, a worker is needed to program the robot that does that job.  Another worker is needed to maintain that robot.

Cloud proponents suggest computing power is moving into the cloud and that might be true from a data center perspective.  But what about a user perspective?  A good argument could be made for saying that computer power is moving into users' hands--literally.  Take one trip to the airport and you'll see more and more people using some sort of handheld computer, such as cell phones, smart phones, PDAs, laptops, media players, etc.  These devices have rapidly improving technology, increasing capabilities and increasing usage.  As the possibilities with handheld devices continues to grow, so will users' sense of entitlement and desire for empowerment.  This goes beyond the employee-owned PC concept; this is an "information anywhere, anytime" attitude.  Citrix's MarkT has been talking about this for a while now.  SOA ties in with the "any device" scenario, but how will IT adjust to the possibility that users will end up becoming their own "peer-to-peer cloud," much like we've seen with peer-to-peer file sharing?

As the computing landscape evolves with technology, I think IT will be important player vs. a bystander.  I see IT being responsible for managing technology changes by architecting solutions, determining which services can and cannot be provided by the cloud, empowering users while maintaining compliance and protecting valuable information, etc.  

Cancel

-ADS BY GOOGLE

SearchVirtualDesktop

SearchEnterpriseDesktop

SearchServerVirtualization

SearchVMware

Close