This is where things get muddy, or should I say cloudy?  The company that I work for provides a multi-tenant cloud for Federal use only.  So, from the start, it’s not truly public, since the public does not have access to it.  While different agencies may have virtual machines residing on the same cluster, non federal agencies will not.  Again, I think the reasons are valid and obvious.  Now let’s look at self-service.  By the definition given in the VSP training this has two requirements.  One is you can purchase the required virtual components (CPU, memory, storage) for a new virtual machine without interaction from other parties, the other is that you can increase those components on demand.  The training used a vending machine as a metaphor.  As you can imagine, this isn’t possible with the Federal Government.  There are strict procedures that must be adhered to when purchasing and procuring equipment or services.  This slows the process down, and is at times frustrating for us vendors, but it is your money they are spending and given the whole debt ceiling,  government spending debate. going slow is not a bad thing.  But for the government, purchasing cloud services is less like a vending machine and more like, well like standard government procurement.

Let’s take self service one step further.  Self service is not only purchase on demand, it’s also provision on demand.  This is possible in a Federal cloud, if you plan for it.  You can purchase a set of resources that you pay for at a reoccuring monthly rate.  You could then use these resources as you see fit.  So, if you purchase more memory than you need, more CPU than you need, more storage than you need, you could create and destroy virtual machines as often as you want.  Need to add memory to a VM?  As long as you have some lying around, go right ahead.  Or maybe you want to free-up unused memory on one VM and give it to another, be our guest.  Resource pools can be wonderful things.

So it’s not a public cloud, it’s not a private cloud, it’s not a hybrid cloud.  Above that, it’s not built for self service, at least not fom the purchasing aspect.  So by VMware’s definition, it isn’t even a cloud.  So what is it?  I’ve heard the term “Community Cloud” floated around.  That’s probably as good a description as any, if you think of government agencies as a community.  Of course, as with all communities, there are going to be sections of the community that don’t want to associate with others in the community.  In the case of the community cloud, this will be due to security reasons.  Systems rated low, medium and high will most likely need to be kept separate, and then of course there is Defense and Intelligence that like to stay off to themselves, and not associate with anyone else, but you have those people in your community don’t you?  If you can’t think of anyone, then it’s probably you

So is it a cloud?  If so what type?  Should VMware and the virtual community shift the way we think of clouds to accomadate federal requirements or is there a fourth type of cloud?  Surely the United States government isn’t the only government that has these concerns and strictures.  Feel free to weigh in.

That’s the question I asked myself when I came across an intriguing job posting.  The answer could be found on the company’s website.

The Eyak people originated along the Eastern Coast of Prince William Sound, where the town of Cordova, Alaska is located. With less than 400 Eyak Natives, it is the smallest Native group in the state. Traditionally, they fished and hunted in the Copper River Delta and along the forests and coastline that runs along Prince Williams Sound. The people lived between the Athabascan Indians to the north and west, and the Tlingit to the east. Aleut and Chugach (Alutiiq) people also lived along the western shores of Prince William Sound.  While retaining their own unique identity and language, the Eyaks traded and adapted certain customs and tools from their neighbors.  Like the Tlingits, the Eyaks are socially organized into a clan system of the Eagle and Raven….Many contemporary Eyaks live in or near Cordova, on the Copper River Delta. It is an area rich with abundant wildlife and breathtaking scenery.

Ok, that answers that, but what does that have to with technology?  Back to the website……

Eyak Technology, LLC (EyakTek), an Alaska Native-owned Small Business, provides award-winning solutions in infrastructure and security systems, communications, information technology, and healthcare services. As a recognized leader in the industry, EyakTek has consistently been ranked as one of the Top 100  Federal Contractors.

Well, if you get beyond the marketing fluff, the last four words are promising.  But why are they looking for a senior virtualization engineer?  What is their involvement with virtualization?  Website again.  I must say, they do have a very easy to use and informative website.

EyakTek assists in all areas of the Virtualized Environment’s Lifecycle including Assessment, Design, Project Management, Installation, and Maintenance.

We start by providing virtualization assessments that examine your current environment to see where cost savings can be realized. We can help prepare a TCO study and report so that you can see where savings exist, and where we can support you in the design phase and installation.
 
In addition, we have extensive project management experience to ensure that your virtualization projects run smoothly and meet deadlines, cost expectations, and achieve your mission.

With our wide-ranging partner relationships with leading technology companies, we can assist in the supply and acquisition of all the supporting hardware and software that best suits your virtualization needs, providing the right virtualization tools that are right for your environment.

Well that contains a lot of marketing fluff too, but it does contain some promising buzz words.  But there has to be more.  You guessed it, website time…..

As a component of our diversified government business, EyakTek is focused on Cloud Computing as a core capability, including the associated technologies of virtualization and managed cloud services. From virtualizing resources in an existing data center to providing a utility-based computing platform, we have the in-house expertise and the partnerships necessary to deliver IT as a complete and managed service. EyakTek can facilitate a seamless transition to a scalable, customized cloud platform structured to fit any environment….

EyakTek is committed to cloud-based services delivering high value, efficient technology solutions, flexibility, and ease of use to customers. Both the IaaS and Data.gov BPAs offer cloud-based services to agencies that deliver high-value, efficient technology solutions via a procurement methodology that provides flexibility and ease to customers. Combined, the two BPAs dramatically expand EyakTek’s ability to deliver a wide array of cloud-based hosting and computing services to all federal, state and local agencies.

Then I came across these articles on, you guessed it, the website……

Eyak Technology LLC, (EyakTek) was recently one of five awardees on a General Services Administration (GSA) 5 year, $46.4 million Blanket Purchase Agreement (BPA) for shared services and dataset hosting for Data.gov and the agencies it serves. Under this BPA, GSA offers expanded hosting and related services for agencies that otherwise cannot host datasets for public use, or for those agencies wanting to take advantage of the shared services environment.  While the BPA covers a broad range of services, it includes data storage, format translations, and download environments.

AND

Eyak Technology, LLC (“EyakTek”) announced today that the company is one of eleven awardees of a Blanket Purchase Agreement (BPA) from the General Services Administration (GSA) for cloud computing. The 5 year, competitively bid Infrastructure as a Service (IaaS) BPA is valued at $76.5 million. To meet GSA’s need for hosting services and solutions, EyakTek has teamed with Horizon Data Center Solutions (HDCS) to provide cloud-based IaaS offerings for customers that will be available through Apps.gov, the government’s cloud-based services storefront. 

Services Available under the IaaS BPA: Cloud Storage Services , Virtual Machines, and Cloud Web Hosting.

Those of you associated with government contracting will realize what a big deal it is to be awarded those BPAs.  For those of you on the commercial side, let’s just say it makes winning government contracts a lot easier.

 I just realized that this post has gotten lengthy.  So to wrap things up, all of this info made me curious to talk to the people at Eyak Technologies.  While not as large as the Lockheeds, CSCs and such, their dedication to Cloud Services and VMware impressed me.  Since starting the job a week ago, I have started consulting with a DoD customer on a couple of data center projects, I have been engaged by a government agency to partner with VMware for a View proof of concert, and we have started responding to the next BPA ….email as a service, using Zimbra.  As you can tell, like me, Eyak is heavily into all things VMware.  With the Federal government’s Cloud First mandate and EyakTek’s zeal for cloud and virtualization, I think this is going to be an exciting place to be.  This is my first post in a long time, as I waited for things to settle down.  Now that I am once again actively engaged in virtualization, I plan on posting much more.  I already have a few things in mind, which is much better than the past.

To quote Microsoft, “To the cloud….”

]]> http://tonywilburn.wordpress.com/2011/05/18/my-new-gig/feed/ 0 tonywilburn http://tonywilburn.wordpress.com/2010/09/15/vmworld-2010-impressions/ http://tonywilburn.wordpress.com/2010/09/15/vmworld-2010-impressions/#comments Wed, 15 Sep 2010 19:44:37 +0000 tonywilburn http://tonywilburn.wordpress.com/?p=168 ]]> I apologize for not blogging more, life has been hectic since changing jobs.  I did get away for my annual pilgrimage to VMworld.  This is my 5th year in a row, and I have to say, I was not very impressed.  Much of my frustration has to do with the sessions.  Or more accurately, my inability to get into sessions. 

In years past, there has been much complaining by people that signed up for VMware late, about not being able to register for sessions and having to stand in an unregistered line and hope they can get in.  So this year, no registration for sessions.  This has led to incredibly long lines and sessions closing a full 15 minutes before scheduled start.  But, of course, the session doesn’t start early just because the doors are closed, if you were lucky enough to get in, you have to sit there for as much as 30 minutes waiting, checking email, blogging, tweeting, etc.  Personally I was denied five sessions that I tried to get into, that is before I gave up trying.  I will admit, later in the week, at least the organization improved as the lines were better defined and ushers were posted at the end of the line with signs that had the room number so that you can see where you needed to be.  Some sections had ushers that had learned how many people could fit in a room and could tell you if you had a shot of getting into a session by the length of the line.  So kudos to the support staff.

I’m not sure why this bugged me so much as I often leave sessions filling unfulfilled.  I guess I have been doing this so long that I expect more content.  Instead I sit through a session of people telling me what I already know and presenting it as something that they newly discovered.  Either that, or it turns out to be, not a technical session, but a sales session, with a vendor telling you why you need their product.  I’m okay with that, but let me know that it’s a sales pitch up front so I can make a better informed decision.  I will probably stick with Advanced sessions only from this point on.

Speaking of this, social media has changed the way VMware sessions are booked.  With the heavy usage of blogs and twitter by the VMworld community, some sessions were in high demand.  I think Scott Lowe’s session was full over 30 minutes ahead of time. 

The long lines for sessions also kept me from seeing as much of the Solutions Exchange as I wanted.  In the past, with pre-registration, after a session I would head to the Solutions Exchange and talk to a vendor or two before checking in at my next session a whole 5 minutes before the session started.  Doing this between every session would allow me to meet with several vendors or network with peers.  Under the new system, it was a mad dash from one session to the next, in hopes that the line was still short enough, thirty minutes before the session started.  A colleague of mine was actually lining up for sessions an hour in advance.  So I didn’t see as many sessions as I wanted, talk to as many vendors as I wanted or meet with as many people that I follow through blogs or twitter as I wanted.

Now, to the good.  Herrod’s presentation was amazing.  I am a huge proponent of VDI and the things that are happening in that space are much needed and highly anticipated.  There were some good session.  Some I saw, some I didn’t get to in time.  There were some very good vendor presentations in the Solutions Exchange.  I think there was something there for everyone and a lot of VDI enabling products to see.  The labs were great.  I should have spent  much more time there.  While there is only so much you can do in the alloted time, especially if you stick with the script, they did an amazing job in the lab.  The labs were actually hosted “in the Cloud” as Herndon, VA and Tampa FL as well as in the Moscone Center.  I only did a few labs, but they were very responsive.

I was able to do some networking, meet up with new friends and old.  My company has several attendees here from literally all over and I was able to have dinner with some old friends and make some new friends within the company.  I am sorry I missed seeing so many people that I had hoped to see.

Now, I’ve never had to plan an event larger than a kids birthday party, so I can’t imagine all the hard work that the VMware staff put into this, and yet, I can’t help filling underwhelmed.  The impressions of many is that the Moscone Center isn’t large enough, even with the addition of the West Hall, for 17,000 people.  But, I have been told, that the Oracle Openworld conference is over twice the size and held in the same conference center.  Are the attendees there jammed together or is it somehow organized to handle such a large capacity?

What I can tell you is that VMware is listening to it’s people.  Many people expressed displeasure that the event was not held in Vegas, and next year, it’s in Vegas.  Many people have expressed a dissatisfaction with the cold sandwich lunches provided.  This year there were some very good hot lunches to be had.  Many people have expressed dissatisfaction with the scheduling of sessions and labs.  This year it was first come, first served.  VMware is continually trying to improve VMworld just like they continue to improve their products.  And like changes in their products, some I like, some I am against. 

Even though there were somethings I didn’t care for, it was still a positive experience, and I look forward to attending VMworld 8.0 in Vegas next year.

So what does this have to do with technology?  Well, I think it’s the perfect example of what not to do when faced with a problem, technological or otherwise.  Many times I’ve seen colleagues fail to correctly troubleshoot an issue because they hadn’t looked at any documentation beforehand.  RTFM as they say.  Her second mistake, another classic, she blamed the computer for her failings.  People seem to believe that computers have a mind of thier own or able to things other than what they were told.  She gave the computer the wrong instructions and expected it to do what she wanted, not what she told it.  Third, she paniced.  Instead of slowing down and trying to figure out why what she had been doing hadn’t worked, she immediately tried another solution.  Fourth, why didn’t she try the most obvious “fix” first.  If she had a dollar bill or four quarters, why didn’t she use them to begin with?  Instead she tried the most expedient solution, even if it wasn’t the best solution.

So what should she should have done?  What should you do when you are troubleshooting an issue?

1.  Make sure you know what you should be doing before you do it.  RTFM.

2.  Take your time.  When you panic, when you rush things, you make mistakes that cost you more time than it would if you had slowed down and done things properly the first time.  I know that no one can work, the phones are ringing off the hook and the CIO is screaming to know who is at fault.  Don’t worry about all that, take a deep breath and concentrate on the task at hand.  You can fix this.  You know you can.

3.  Ask questions.  Don’t be afraid of looking stupid.  Face it, if she had asked me for help the first or second time she had an issue, I would never have written this blog entry.

4.  The easiest fix, isn’t always the best.  Take the time to look at all available solutions before choosing your course of action.

Four simple steps, I’m sure you could add more, but if you follow these rules, your endeavors have a much greater chance of being successful.

NOTE:  This must be done for every Initiator Group on each SAN Controller.

Launch NetApp System Manager.  Choose LUNs in the left pane.  Choose the Initiator Groups tab in the right pane.  Click on Edit and Check the box to enable ALUA.

Log on to the vCenter Server.  Copy the latest NetApp VSC to the vCenter Server.  Launch the executable.  Setup is an easy and straight forward wizard.  I won’t go into the details of it here.  Once the VSC is installed you will be prompted to register it.

Launch vCenter Client from a workstation.  You will receive a security warning when logging into vCenter Client.  Check the Install box and click Ignore.

You will now have a NetApp tab.  Choose the NetApp tab to access the plugin.

The plugin will begin to discover your network components.

You will need to provide the credentials to log in to your controller.

NetApp will show a summary of your SAN.  Click Finish.

If the MPIO settings show alert you will need to let NetApp adjust the settings.

Right click on the host and click on Set Recommended Values.

Only the MPIO settings are necessary, but I leave all the boxes checked, just to get Green lights on all the settings.

The Status will now show Pending Reboot.

Reboot the vSphere host.  After the host is running again Update the status.

After the update the MPIO should show normal.  If the Status still shows alert, I find that if you wait a couple of hours, it will turn to green.  Not sure why there is a delay in the status update.

The Multipath settings should now show Round Robin, with two paths active.

That’s it.  Just do this for every vSphere host that you have and you will have 2 active paths for every LUN.  Much better than using defaults or changing the settings for every LUN on every host.

I never imagined that two little sentences would be posted and quoted and re-quoted as much this one remark has been over the past few days.  So let me delve into this a little more and explain why I said what I did.  Before I got into virtualization, I was a Novell Engineer.  I was so well versed in Novell products that for awhile, I was a Novell Field Engineer, or in Novell’s terms a Designated Support Engineer.  Novell, in many people’s opinions,  had a superior product, yet NetWare and more importantly Novell Directory Services (NDS) has been crushed by Microsoft and Active Directory.  Some people would argue that Lotus Notes and GroupWise at one time were better products than Exchange, yet Exchange now owns the corporate email market.  Most of us old-timers will expound to great lengths about how great Word Perfect was, and now Word Perfect wouldn’t even be able to get on Celebrity Apprentice.  Then, of course, there was Netscape.  You remember Netscape, the ultimate in browser experience, don’t you?  How did Microsoft overpower all of these arguably superior products?  Through integration.  Active Directory was bundled with Windows Server, just as Internet Explorer was bundled with every Windows version since Windows 95.  Now Microsoft’s latest target is VMware.  How have they attacked VMware, why integration, of course.  Hyper-V integrated with Windows 2008 and 2008 R2 to be exact. For more years than I can remember, Microsoft has owned the desktop space.  Even before there was a Windows product, most applications ran on MS-Dos.  Now, they see VMware as attacking that domination with View.  Why Microsoft sees this as such a challenge is beyond me.  VMware hasn’t designed their own desktop OS, they’ve only virtualized Microsoft’s.  In doing so, they have enabled Microsoft to charge for two licenses instead of one.  More on that later.  Now to the announcement, among other things Microsoft has teamed with Citrix for not only the new RDP protocol called RemoteFx, but also a “trade-in” program which lets customers trade in their VMware View software licenses for the same number (up to 500) of Microsoft VDI Standard Suite subscription and Citrix XenDesktop VDI Edition annual licenses, free of charge.  Granted Microsoft and Citrix have worked together for years, as Citrix was almost a requirement to use Microsoft Terminal Services.  But in none of the cases that I sited above did Microsoft need to partner with another company in order to oust it’s competition.  It’s like when the schoolyard bully finally meets his match and comes back with a friend to try again.  That only makes the bully look weaker and his intended victim even stronger.  There.  I said it again.

Microsoft’s VDI license announcement:

I’ve been trying to decide if I should say what I really feel about this part of the announcement or just shake my head and walk away.  Seems I’m not the walk away type.  I feel this much hyped announcement was very, very weak.  First, we all know that Microsoft practically gives away the OEM copy of Windows XP/Vista/Windows 7, so that it will be pre-installed on your PC and you won’t want to go through the hassle of changing it to anything else.  Yet, if you buy this same software from the store, Microsoft charges over $200.  Windows 7 Professional is $261.76 today on Amazon.  This drives up the cost of any VDI solution, be it VMware View or XenDesktop, making ROI more difficult than it should be.  On top of that Microsoft charges a fee to attach to that desktop from another desktop or thin terminal.  And if that other desktop or thin terminal is running a flavor of Windows, guess what, you are charged for that too.  Now Microsoft has made licensing “easier”.  If you are a large enough customer that you have an SA agreement, you are allowed to connect to that Virtual Desktop from any secondary, non-corporate network devices, such as home PCs and kiosks.   If you do not have an SA, you will need to purchase a VDA license.  Starting July 1st, 2010, customers that intend to use devices that do not qualify for Windows Client SA (such as thin clients or third party contractor PCs), will require a license called Windows Virtual Desktop Access that can be purchased for $100 a year.  The previous VECD license was $110 a year, giving you a savings of $10 for the new licensing.  Also, a little talked about penalty is the license can only be associated with an end point every 90 days.  So, if you have a consultant using his own desktop for one month, that license is unusable for 60 days.  Why is this?  I paid for a license, yet I can’t use it how and when I want?  I’m sure this makes sense to someone out there, but to me, it sounds like I am leasing a car and I can drive that car as much as I want until my wife drives it, then I can’t drive it for 90 days, even if she just drove it to the grocery store and back one afternoon.  This just doesn’t make sense to me.

Another thing I find interesting is Microsoft and Citrix are pushing this “trade-up” plan.  Yet you “trade-up” from VMware View which is built on ESX Enterprise Plus and includes Thinapp to Microsoft VDI Standard Suite.  You’ll have to pay extra for:

1. Complete Remote Desktop Services capability, including the option to deploy session based desktops in addition to VDI desktops.
2. Microsoft Application Virtualization for Remote Desktop Services.

On the Citrix side, you are getting the VDI Edition.  There are two levesl above this, Enterprise and Platinum.  I’ll let you see the breakdown of these below.

So, I look at this as good news, bad news.  VMware “gets” desktop virtualization.  Way back in 2006 when I was doing my first VDI implementation, VMware understood what I was trying to do and worked with me on it.  Microsoft, at that time, did not understand  virtualizing a desktop on a hypervisor, at least the Microsoft employees that I tried to work with didn’t.  Citrix only wanted to talk about virtualizing the applications, which didn’t go far enough to suit my needs.  Now, finally, four years later, Citrix and Microsoft “get it”.  Now if they would only give the customer what they need for a price that makes sense, maybe we could have parity and a true competition that would drive innovation.  This was a step toward that, however a very small, almost grudging step.

And that is what we are getting at here.  I’m not going to do price comparisons with XenServer and Hyper-V, that’s been done to death by all three camps.  I just want to put my views down where I can see them to help me better present my thoughts.  First, sticking with the performance theme.  VMware pays a lot of smart people to develop things that lets face it, their competitors don’t have.   Lockstep Fault Tolerance, Memory Over Commit, Transparent Page Sharing, Storage VMotion, Distributed vSwitch, Host Profiles, and coming soon, Memory Compression.  I’m not going to argue the merits of these and why I think they are invaluable, I’m just going to reiterate, you get what you pay for.

But lets look at how expensive VMware really is.  One thing I have seen is that most cost calculations leave off the fact that primary motivator for the  move to a virtualized platform is to retire old hardware.  Warranties on typical servers are three years, even though I’ve seen many companies stretch the life cycle to five years or more.  I’m going to use a three year refresh rate in my computations.

So, lets say you are a small to midsized business and this year you have 20 servers that are going end of life.  You could purchase 20 new servers to replace them, or you could purchase three new servers plus VMware licenses.  I’m going to use vSphere Enterprise for my example, since a SMB wouldn’t need Enterprise Plus.  The table below shows the “cost” of VMware.  The only thing that isn’t included in this is storage as there are too many variables in storage.  Some people already have the free space on the SAN, some only need to buy disks for an existing SAN and some need a new SAN.  There are too many different scenarios for me to include them all in this short blog entry.

So the “cost” of a VMware solution is a savings of $80,066.  Even if you don’t have the time and resources for the upgrade, you can hire a Solutions Architect (like me, for example) for less than the $80k and still come out ahead.  So, I’m a forward thinking person, I don’t just look at the right now, I like to look at, lets say, five years down the road.  Let’s assume you have a modest 20% growth in your server farm.  What happens over the next five years?

Now year three gets even better because we have plenty of room for the 20% growth with the existing infrastructure.

Year four.  Time for a server refresh and time to purchase SnS for vSphere.

Year 5.  Again Time to refresh the year 2 servers and purchase SNS.  Again we have room for the 20% growth in the existing infrastructure and do not need to buy new servers.

Granted this all assumes that the price of vSphere doesn’t increase and the prices of servers doesn’t increase.  I’m sure some of you noticed I left the 20% static from the first year instead of increasing it every year.  I just found the charts easier to read and understand if the numbers stayed the same.  The difference is only six servers over five years which I think is fairly negligible for this exercise.  I feel this is a more real world exercise than what most analyst show when they compare purchasing VMware with doing nothing.  Most analyst when explaining why VMware is too expensive will show you $60,000 vs $0 in the first example.  They forget that time marches on.

So, where does this leave us over a five year period?  Just in case you weren’t keeping score, we now have thirty-six servers, or thirty-six virtual servers running on five vSphere hosts.  A modest ratio of 7:1 which will allow for a one host failover solution giving you a 9:1 ratio during a host failure or maintenance.

Wait a minute, Tony are you saying that I will save almost $300,000 over five years just virtualizing 20 servers and virtualizing all my new servers?  Yes, that’s what I’m saying.  And not only do you save almost $300,000, you now have increased uptime through vSphere High Availability.  Your System Administrators can now do hardware maintenance and repairs during the day and spend nights and weekends with their families.  This may not seem like much to a manager since the System Administrators are on salary and get paid the same for a ninety hour week as for a forty hour week, but trust me, as an ex-System Administrator, this will increase the moral and decrease turnover.

But wait, there’s more!  Not only does it cut and dice, chop and slice, it’s also GREEN!  Lets look at the power consumption over the same 5 years.  I’m going to use nice round numbers and assume each server is running continuously at 300 Watts.  Of course this will vary, but it’s a good average mark.  The smart folks at Dell use a 2.8 Power Usage Effectiveness rating, so I won’t second guess them and I will use the same.  I will also use the same $.10 per kWh that the fine folks at Dell use in their calculations.

That’s a saving of $87,565 over five years of operating expenses.  Bringing the grand total to a savings of $376,721.  If you have a larger environment the savings just increases.  As I said, this doesn’t include storage cost, but you have to purchase storage no matter solution you choose.

So let’s recap.  The high cost of vSphere for twenty servers is a savings of $379,217 over a five year period.   In the first year alone you get a savings of $80,000 on a $60,000 purchase.  Sounds like vSphere pays for itself in less than twelve months.  A fact that has been proven by people like Forrester, who are better trained than I at making these kinds of calculations.  If VMware drops their price, will innovation stop?  Would VMware be able to stay so far ahead of their competition?  And if the cost of vSphere is a savings of almost $400k for 20 servers, how low do you want them to go?  How greedy are we?  Have we become spoiled with the so called free cost of open source software?  Don’t just look at the license cost, do an in-depth analysis, then decide if you can afford not to virtualize.

As the years passed I discovered that there was something that I was good at.  Computers.  Working with computers I was a natural, PCs, servers, networks, it didn’t matter, I could handle it.  The years passed by and I took on the look of a Systems Administrator.  You know the ones, goatee, what hair I have left cut short, overweight from all the Mountain Dews I drank trying to stay awake doing after hours system upgrades while the users were home living their lives.  Now don’t get me wrong, I know we don’t all look like this, some of my peers actually do things like …gulp… run, and without chasing someone or being chased.  I’m just saying that if you were building the stereotypical Systems Administrator he would look at lot like this….

Imagine my surprise when I saw Steven Holcomb…

I turned to my wife and said, “He missed his calling, he should have been in IT.”  Well, as I watched him over the next several days, he placed sixth in the two man bobsled, and then guided his team to the gold medal in the four man bobsled.  Steven Holcomb definitely did not miss his calling.  But still, I couldn’t help being curious, so I did some research.  I was amazed at what I found….Holcomb majored in Computer Science at the University of Utah and the University of Phoenix’s online program.  He is an A+ Certified Computer Technician, Network+ Certified Network Technician, and a Microsoft Certified Professional.  What?  He’s one of us!!!!  You mean the dream of being an Olympic champion is not out of reach, even for us IT guys?  I find this out now, when I’m 40 years old?!?!?!?  But, I could’ve been a contender.  I think Holkie said it best on twitter, “Find a dream, work for it, then live it!“

Steven Holcomb, thanks for showing us that anyone, even an IT geek, can realize the Olympic dream.  Congratulations on your Gold Medal.