Friday, March 12, 2010

Enterprise Cloud Computing — Build Your Own With Cisco VFrame — Why Wait ?

August 21, 2008 by Greg Ferro · 18 Comments 

I can see some value in external Cloud Computing, but why not just build your own with Cisco ? Take a bunch of leftover machines, that old stor­age sys­tem and get a demo ver­sion and make your own.
It seems clear to me that Cloud Computing is going to go down two quite dif­fer­ent paths. The first is the path that Amazon /​ Google /​ Joyent rep­res­ent. Enough said on those technologies.

But why not build a Cloud Infrastructure in your own Data Centre ?

The secur­ity and oper­a­tional risk of hand­ing your applic­a­tion out to some­thing like Google means its unlikely to hap­pen any time soon. Moreover, rewrit­ing your busi­ness applic­a­tions for their envir­on­ments is prob­ably never going to hap­pen. No mat­ter how I try, SAP on the Google is not going to hap­pen in my lifetime.

The good news is, that Cisco already makes a tool that does, more or less, does a ‘do it your­self’ Cloud Computer. Lets look into that.

Cisco already makes a cloud solution

Sometimes there are so many products in the Cisco cata­log that some of them get lost. But one product I have been research­ing over the last six months is Cisco VFrame.

Now VFrame is soft­ware tool­set that auto­mat­ic­ally pro­vi­sions VMware ESX serv­ers, Cisco Catalyst switches, ACE Application Delivery Controllers1 , FWSM mod­ules, Storage Arrays and Storage Switches and so on. In simple terms, it is pro­gram­ming envir­on­ment that allows to ‘orches­trate’ the con­fig­ur­a­tion of many sep­ar­ate tech­no­lo­gies into a single process.

Let me ay that last bit again ” ‘orches­trate’ the con­fig­ur­a­tion of many sep­ar­ate tech­no­lo­gies into a single pro­cess.” Thats is the first part of Cloud Computing right there.

Looks Like a Cloud

Essentially a tool­set that spawns a vir­tual machine, and all the sup­port­ing net­work infra­struc­ture. VFrame will :

  • auto­mate the cre­ation and deploy­ment of VMware hyper­visors, with all the stand­ards and tool­sets you specify
  • auto­mate the net­work oper­a­tional changes includ­ing fire­wall, load bal­an­cing configuration.
  • auto­mate the cre­ation and alloc­a­tion of stor­age units to the virtual

If you ever cre­ated a sys­tem on Amazon EC2 (their so-​​called elastic com­pute cloud), you get the feel­ing that this is exactly what they have done.

Smells Like a Cloud

Lets look at the bit of Amazon EC2 :

To use Amazon EC2, you simply:

Create an Amazon Machine Image (AMI) con­tain­ing your applic­a­tions, lib­rar­ies, data and asso­ci­ated con­fig­ur­a­tion set­tings. Or use pre-​​configured, tem­plated images to get up and run­ning immediately.

Upload the AMI into Amazon S3. Amazon EC2 provides tools that make stor­ing the AMI simple.

Amazon S3 provides a safe, reli­able and fast repos­it­ory to store your images.

Use Amazon EC2 web ser­vice to con­fig­ure secur­ity and net­work access.

Choose which instance type(s) you want, then start, ter­min­ate, and mon­itor as many instances of your AMI as needed, using the web ser­vice APIs.

Determine whether you want to run in mul­tiple loc­a­tions, util­ize static IP end­points, or attach per­sist­ent block stor­age to your instances.

Pay only for the resources that you actu­ally con­sume, like instance-​​hours or data transfer.

So if I have Cisco VFrame, a soft­ware tool that auto­mates the fol­low­ing tasks:

  • alloc­ate and cre­ate a stor­age capa­city and stor­age net­work from my exist­ing stor­age system
  • alloc­ate and cre­ate CPU /​ RAM resources in the form of a VMWare ESX hypervisor
  • cre­ates the net­work modi­fic­a­tions for fire­walling and applic­a­tion acceleration

then that smells lot like Cloud Computing to me.

Feels like Cloud Computer.

So the VFrame soft­ware is able to con­nect to the Storage Array using vari­ous APIs, then cre­ate stor­age con­fig­ur­a­tion and trans­fer the ESX image to the LUN.

The stor­age net­work is then con­figured to add the LUN’s to the MDS units.

The net­work is con­figured for IP address­ing, switch ports, NIC team­ing and any other switch con­fig­ur­a­tion. Automated con­fig­ur­a­tion for FWSM for secur­ity, and ACE for load bal­an­cing and applic­a­tion acceleration.

VI3 Remote Boot is used to start the ESX server.

Given that this all auto­mated, we can use our stand­ard build pro­cesses for every machine that is cre­ated, and the ongo­ing main­ten­ance of these sys­tems becomes simplified.

Elastic Computing

OK, where is the ‘elastic nature’ of the Cloud. The elasti­city comes from the under­ly­ing nature of the technologies — 

  • VMotion takes care of scal­ing CPU and RAM resources from a pos­sible pool of resources
  • dynamic stor­age alloc­a­tion on your MDS switch (com­bined with your Storage Arrays) provides scal­ing of stor­age capacity
  • net­work vir­tu­al­iz­a­tion allows you to move fire­wall and load bal­an­cing within the net­work to find band­width and pro­cessing resources.

Given that I can do this already, why am I out­sourcing this to the Google or the Amazon.

Outsourcing usu­ally fails

While I have seen out­sourcing be suc­cess­ful, it almost never is. Outsourcing the com­puter func­tions of your IT may be an OK idea for mail other simple func­tion like Office, it’s hard to per­ceive that out­sourcing core com­pute func­tions will ever happen.

But we need some­thing Cloud Computing in the change phase of our Data Centres. Funny thing, it’s already here.

Must be a Cloud!

I am sure I haven’t done this justice, and you should head over to Cisco’s Web site and get a look at the mar­ket­ing mater­ial at www​.cisco​.com/go/vframe and get the full picture.

Here is thing, most people can build a test envir­on­ment for a lot of this with the stuff you are throw­ing away. Get some of those old serv­ers and stor­age arrays and build a test envir­on­ment. Imagine how cool it would be to have your own “Cloud Computer”. Now that would be brilliant.

The thing I don’t under­stand, if why Cisco is not giv­ing this more focus. They should have their entire “Social Marketing” machine out their push­ing this like their lives depended on it.

There are some really amaz­ing tech­no­lo­gies in Cisco’s Data Centre strategy, and I reckon that a lot of people don’t know about them. This is just one of them. I have some more tech­nical stuff com­ing in the next few on net­work infra­struc­ture, so keep any eye.

Thanks for read­ing this far, feel free to com­ment and add inform­a­tion. Cisco Web site on Cisco VFrame is here.

Footnotes

  1. Application Delivery Controllers — yep, that the new term for a load bal­an­cer. Well, ok, they mimght do some applic­a­tion accel­er­a­tion and app fire­walling, but it is still a load bal­an­cer. [back]

Please rate this post:

  Why Rate Posts?
1 Star - It\\\'s Crud2 Stars - It\\\'s Tosh3 Stars - Something\\\'s missing4 Stars - Needs works5 Stars - Good Enough6 Stars - Good7 Stars - Excellent8 Stars - Brilliant9 Stars - Astonishing10 Stars - Awesomely Godlike? (No Ratings Yet)
Loading ... Loading ...

Comments

18 Responses to “Enterprise Cloud Computing — Build Your Own With Cisco VFrame — Why Wait ?”
  1. Sounds Cool! The Data Center stuff is Insane. Maybe there will be a new CCIE Data Center. Seems like the trend. More to study for. My head hurts!

  2. …or one could use Scalent V/​OE, or Unisys uAdapt…

    Check out slide #5 in Network World’s “10 Must-​​have Virtualization Tools”

    http://​www​.net​work​world​.com/​s​l​i​d​e​s​h​o​w​s​/​2​0​0​8​/​0​8​1​8​0​8​-​n​d​c​-​v​i​r​t​u​a​l​i​z​a​t​i​o​n​-​tools.html

    (Yes, shame­less plug!)

    • Greg Ferro says:

      Yeah, but its in an art­icle called vir­tu­al­isa­tion tools. All the buzz is about “cloud com­put­ing”, which is of course, the new mar­ket­ing term for vir­tu­al­isa­tion 2.0.

      I have never heard of Scalent so I guess you don’t have much of pres­ence out­side the US. Let me know when you get global.

      No prob­lem with plug, dis­clos­ure is the key to mak­ing it acceptable.

  3. …shame­less plug continued…

    Yeah, we’re not yet quite the size of Cisco. Working on that.

    But we’ve actu­ally got quite a reas­on­able EMEA pres­ence, dir­ect and through vari­ous resellers — and recog­ni­tion, see here:
    http://​awards​.tech​world​.com/​w​i​n​n​e​r​s​2​0​0​8​.​a​s​p​?​m​_​p​i​d​=​0​&​a​m​p​;​m​_nid=29322 (“Virtualisation Product of the Year”)

    EMEA con­tact phone num­ber on our site…

  4. Greg,

    I liked your post and wanted to let you know that I ref­er­enced you and your post in a Cloud Computing Examples post today. My thought is that a lot of the lar­ger organ­iz­a­tions with net­work and server sup­port staffs will look at build­ing their own Cloud Infrastructure as opposed to mov­ing their data into the cloud.

    Thanks for the post  —  Kevin Mullins

  5. Bill Erdman says:

    Greg,

    Thanks for your insight­ful post­ing on build­ing ad-​​hoc com­pute based clouds with Cisco’s VFrame Data Center offer­ing. This product does have a sig­ni­fic­ant devel­op­ment team behind it with good mar­ket­ing mater­ial as you have read. While it may not appear that Cisco is heav­ily selling this product, Cisco is work­ing with sev­eral large enter­prise and ser­vice pro­viders cus­tom­ers on the adop­tion of this technology.

    As you spec­u­lated this is a non tra­di­tional product offer­ing from Cisco and we are learn­ing how to pos­i­tion and sell this to cus­tom­ers. There is a lot of device integ­ra­tion required when you think about tying the pro­vi­sion­ing of serv­ers, net­work and stor­age devices together as a set of mal­le­able, vir­tu­al­ized cloud resources. Every week we add more devices to the VFrame DC com­pat­ab­il­ity mat­riix. As you stated we encour­age cus­tom­ers to try this product.

    Thanks for vist­ing our web site.

    Bill Erdman, Senior Director, Cisco Data Center Technology Group.

    • Greg Ferro says:

      Dear Cisco Marketing

      See that, yep, that com­ment just above this, now _​thats_​ Social Networking.

      Idiot com­ments on twit­ter, and poncy wiki’s don’t count diddly.

  6. Interesting art­icle (and posts) about a topic that’s get­ting a lot of buzz in the mar­ket these days. I think Greg’s got it right as far as he goes, but there’s more to cloud than tech­no­logy virtualization…one of the most inter­est­ing aspects of cloud is the fact that it also vir­tu­al­izes geo­graphy. Not only can tech­no­lo­gies be toggled on and off-​​line to meet busi­ness demands, but the under­ly­ing require­ments for applic­a­tion per­form­ance (tied to latency and jit­ter), reli­ab­il­ity (tied to phys­ical redund­ancy), and data integ­rity (tied to data base rep­lic­a­tion) can also be achieved by link­ing mul­tiple data cen­ter facil­it­ies over a shared large-​​scale net­work, all con­trolled through a com­mon portal.

    One final comment…the exist­ence of a global out­sourcing industry would seem to refute the com­ment that out­sourcing gen­er­ally doesn’t work. The only way to scale the tech­no­lo­gies that Greg’s work­ing with is through large infu­sions of cap­ital which can either be provided by the enter­prise or by a ser­vice pro­vider who shares that invest­ment across enter­prises. There is an entire industry serving the needs of enter­prises that is thriv­ing between the DIY IT man­ager and the likes of Amazon/​Google.

    Great stuff!

    • Greg Ferro says:

      The exist­ence of a global out­sourcing shows the lim­ited cre­ativ­ity of senior busi­ness exec­ut­ives. Outsourcing was very fash­ion­able and many com­pan­ies now regret that decision. Increasingly the news tells of out­sourcing con­tracts being reversed.

      This is no acci­dent. The busi­ness goals of an out­sourcer are not aligned with cus­tomer and usu­ally causes fric­tion. Its an unhappy affair at best.

      One day I might find a per­sonal exper­i­ence of a happy out­source, but it has not happened in the last seven years.

      The external host­ing cloud thing is tem­por­ary. The concept of Amazon/​Google clouds will suit some people (mostly low cost star­tups mak­ing a lot of smoke but very little fire), but it isn’t going to change the world any­time soon. It’s hype today, some deploy­ment and then.…. just another IT choice.

  7. Carly says:

    After Cisco’s announce­ment of the Nexus 1000V soft­ware switch at VMWorld yes­ter­day, now does it make sense why VFrame was not front and cen­ter, until now? VFrame will be an integ­ral part of the “auto­ma­tion” por­tion of Cisco’s Data Center 3.0 vis­ion, which is com­ing to fruition with the Nexus 1000V, Nexus 5000, and VN-​​Link. Check out VMware’s announce­ments this week as well!

    • Greg Ferro says:

      Since you speak like a mar­ket­ing per­son, I will assume that you are. You haven’t dis­closed any interest.

      Sophistry on this scale must be addressed.

      “now does it make sense why VFrame was not front and cen­ter,” No. In fact, I am will­ing to bet that the VFrame product man­age­ment wasn’t hold­ing back either and try­ing to sell their little hearts out to an audi­ence that didn’t care. Cisco selling VFrame made no sense in the cur­rent mar­ket, after all Cisco didn’t “do” VMWare and cus­tom­ers didn’t believe it. VFrame hasn’t been “front and cen­ter” because it was unsellable. Thats why it wasn’t front and centre.

      If VFrame is an integ­ral part of DC3.0, it hasn’t been announced. Oh sure, I can see the vis­ion, but no fire there. Smoke all you like though.

      Nexus (in any form) is still 12 to 24 months from prac­tical matur­ity. Measure my excite­ment in a years time.

      • Carly says:

        Ouch.
        1. I’m an engin­eer with a com­puter sci­ence back­ground.
        2. I’m 23 and rel­at­ively new to the field, but I am focused on the data cen­ter.
        3. I loved this art­icle…
        Ouch.

  8. Erin says:

    My lim­ited know­ledge and exper­i­ence pre­vents me from join­ing in the tech banter, though in due time, I will enjoy play­ing the devil’s advoc­ate with such top­ics, not to instig­ate frus­tra­tion or irrit­a­tion, purely to advance my under­stand­ing in vastly con­cep­tual mat­ters.
    I am cur­rently doing a quasi RFP for a Networking class in which I selec­ted Cloud Storage as my focus. All the recent hype of EC2, web 2.0, VDC-​​OS, and now the recent crash of the Stock Market poten­tially encour­aging small, medium, and Enterprise busi­nesses to util­ize the Cloud Computing and Storage ser­vices to increase IRR and ROI. I find it amus­ing and a bit sus­pi­cious that the recent stock mar­ket crash par­al­lels so timely with the influx of Cloud Computing.
    Maybe it is due to my research; pos­sibly it is Google Chrome and the sim­pli­fied cor­rel­a­tion that the pub­lic has raised aware­ness. Either way, from an Enterprise busi­ness per­spect­ive, I am grate­ful there are options such as Cisco’s VFrame Data Center and VMware’s VDC-​​OS.

    Greg, I per­son­ally feel that Cloud com­put­ing and cloud stor­age via ser­vice pro­viders can poten­tially be just as hor­rific as util­iz­ing out­sourced vendors in terms of the com­pany data being in someone else’s hands.

    Dennis, you nailed it on the head when you said that “the cloud vir­tu­al­izes geography…underlying require­ments for applic­a­tion per­form­ance (tied to latency and jit­ter), reli­ab­il­ity (tied to phys­ical redund­ancy), and data integ­rity (tied to data base rep­lic­a­tion) can also be achieved by link­ing mul­tiple data cen­ter facil­it­ies over a shared large-​​scale net­work, all con­trolled through a com­mon portal.” This is one of my key factors in selling my quasi-​​proposal.

    Thanks for all your com­ments, and Greg, thanks for your art­icle… I agree, but will someday dis­agree, and not just to ensure that you feel right. :)

    Erin

    • Greg Ferro says:

      Frankly, Service Provider or fully hos­ted Cloud Computing is about as likely and Britney Spears liv­ing a nor­mal life. It seems pos­sible, it should be pos­sible but it won’t hap­pen. This is the third out­ing of the Cloud Computing idea in the last 15 years and it still won’t be suc­cess­ful because the tools don’t exist that make it pos­sible to identify and man­age the blame points.

      However, Enterprise Misty Computing is a def­in­ite, and all of Cisco’s recent moves are for that mar­ket. Of course, if a few ser­vice pro­viders spend some bucks as well thats a bonus but it isn’t the endpoint.

      Service Providers are big, slow, stu­pid and low profit. Enterprises are big, fast, clever and prof­it­able. Which mar­ket would you attack ?

Trackbacks

Check out what others are saying about this post...
  1. […] they will also announce a new ver­sion vFrame, vFrame 3.0 that will fully auto­mated pro­vi­sion of vir­tual machines, fire­walls, load bal­an­cers. and the entire […]



Speak Your Mind

Tell us what you're thinking...
and oh, if you want a pic to show with your comment, go get a gravatar!