TechEd 2014, my perspective

posted in: IT, Uncategorized | 0

TechEd 2014 was a far cry different than any other TechEd I’ve been to. This was my 6th year attending and I have seen many technologies and themes come and go.  This year wasn’t a fad theme based event.  No cool flashy big data demos using features no one will use.  No pimping of “cloud” tech that is largely irrelevant or temporary (ahhem… VM roles in Azure pre IaaS).   What was talked  about and shown for the IT infrastructure and cloud side was not only relevant, but was one of the most confident, and consistent messages I have seen from MS.

Microsoft Azure – its everything.

One consistent theme that has been building like a snowball every year is Azure.   First couple years I attended, Azure was a web platform, and MS’s infantile cloud strategy.  As the years went on, it was a low end VM platform, then an Infrastructure as a Service platform.   This year the easiest way to frame how MS is looking at Azure is “its everything”.

Microsoft is not playing around either.   They are playing with numbers that boggle the mind.  In addition to the current datacenters they already run, they are building (currently, already in production) 20 new DCs at the cost of over 1B$ each, and each takes 3 years to complete. More are planned.   MS has already 1 Million+ physical servers running.  Making them 2nd to only AWS… that is until their new DCs come online. ( )   I would guess only one or two other companies in the world could drop that kind of money up front.  (Apple doesn’t count, because they spend all their money on purely client focused DCs, and their DC scale overall is fractions of the size of AWS/Google/MS)

MS is also the only major player in the cloud market that has released their Hypervisor to the public (Hyper-V 2012), released their management tools to the public (System Center and Azure Pack), and is participating in the Open Compute Project and has published their server specs and designs.

The other thing that was striking to me was how complete the hybrid (On-Prem/In-Cloud) vision is becoming.   MS doesn’t want to just say “everything in the cloud”, because they even know that’s not likely in the next 10 years for most companies.    The Hybrid strategies are clearly geared at making the move to full “cloud” easy, but they don’t force the issue.   Instead they augment existing implementations.  The message was an enhancement on the “Cloud on your terms” theme from the last few years.

The Ultimatum for IT Pros:

MS has thrown down the gauntlet in terms of your career in IT.   If you work in infrastructure, and you shun cloud technologies, you should probably start reviewing your long term job prospects.   Because you can push back, but you will be about as valuable as a Novell Admin in 2014.

 VMware and the competition:

Looking at the real game that is being played shows an interesting story.  MS isn’t fighting with VMware anymore.  VMware’s cloud stragtegy is based on small DCs (relatively) operated by 3rd parties, and they lack the capital to do anything like what MS or AWS are doing.

Frankly, everyone I talked to at MS and even people at the VMware booth were on the same page.  VMware’s diminishing grip on the on-prem virtualization market, and their increasingly out gunned cloud concept, is going to relegate everyone but 2 companies to the niche market.  (which will still be hugely profitable)

The real competitors that MS is focused on is AWS and Google. On-Prem is the battle ground of the past,  consistent on-prem and in-cloud services are the future, and MS is betting everything on it. To the point Mark Russinovich made the comment that “One catastrophic failure of Azure (data leaking, core security failure, etc…) could spell the end of not only the cloud push for everyone, but could end MS as a company. "   Those are some large dice to roll.  In that same interview it became clear Microsoft knows the risks, and knows what is important to focus on.

Money… or lack of conversation about it:

In the keynote, and any and every breakout session you were shown and told how incredible Azure IaaS, DRaaS, etc… are.  The one thing that they didn’t quite cover in many cases is what the reasonable expectation of cost is with some of these implementations.    Storage is getting cheaper, much cheaper, in the cloud.  Connectivity is getting cheaper. and your licensing may become cheaper in a move to the cloud.

But now you are paying for what you use, and that in some cases could be considerably more expensive than you may realize.   Items like the new MASR (Microsoft Azure Site Recovery) will be an incredible feature for the enterprise DR market, but at an incredible cost for lots of VMs.   Currently the Azure  DR orchestration portion alone costs $8/VM per month.   With MASR expect that to be multiplied many times over for storage size, and demand resource that are held in reserve for your DR implementation.

Before a move to Azure in any large capacity is executed there needs to be a detailed look at what you are doing, how you are doing it, and how you can do it more efficiently.   This is where consultants and managed services providers will come in to the mix in a big way in the coming years.


Overall I am encouraged, but at the same time my eyes are open that as an IT Pro I need to consistently refine and grow my skills to stay relevant.  IT infrastructure has had a period of relatively stable progress with skill sets transferring from one project or system to another.   That all changes, and we as IT professionals are about to enter one of the largest shake ups in how we do, what we do.  This is equivalent the move from mainframes to discrete servers.  How many big iron / mainframe admins you see around your IT circles these days?

On the flip side, those who adapt, will have endless work ahead of them in this move.  Yes less IT people will be required (and that’s a key goal of virtualization, and the "cloud”.), but it’s your job to be ready and relevant to the technology you want to be focused on.   There will be plenty of dead weight cut from IT staffs… and that’s OK.   Don’t be dead weight 🙂

Follow Joe Kelly:

Geek of many things

CEO of this Blog. It's kind of a big deal.

Latest posts from

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.