Recent Twitter

IBM v. AWS: Who 'wins' at the Cloud Game?

This morning, I encountered an article by Rob Enderle of entitled: Why IBM Will Win the War with Amazon Web Services.

While I don't believe Enderle takes the time to play this out, there's reason for giving this notion some real, serious consideration.  The unstated, but important premise on which Enderle's story depends is: IBM has used the past three years to watch the cloud industry, sees where it's going, and has bought into the 'open technology' aspects of the cloud industry (in a way analogous to the way it embraced TCP/IP-oriented networking and the Linux operating system). 

Having recognized this, my key take-aways are:

  • Premise: IBM's offerings under the SmartCloud Enterprise umbrella have been placeholders and holding actions, which permits IBM to assess the winning technical directions and offerings, assess their maturity, throw significant resources into development and acqisitions, and emerge with a formidable, contemporary and commercially mature set of cloud offerings. 
  • First, IBM has and knows how to establish credibility with those parts of the organization that make technology purchasing decisions.  Few organizations have EVER out-marketed IBM. And, while we've become accustomed to paying the price for IBM's customization and customer-centric support, those aspects of their products DO impact the purchasing and operating decisions of Enterprise and even SMB IT.
  • Second, so long as IBM fields cloud infrastructure and platform that is performant, compliant and adheres to the standards on which the developer and operations communities rely, the battle's outcome is likely to depend most heavily on credibility, positioning, packaging and attendance to the 'customer facing' aspects.

The Unstated Premise.  Unlike, I would have made the premise much more concrete and factual: IBM's initial push to respond to the rapid (... ok, the explosive) growth of cloud infrastructure services was a 'holding action.'  It has been, in great measure, a means by which a large (and entrenched) IT company responds to a market demand that AWS has both created and then served so well.  In this regard, I view IBM's recent activities in revising their cloud computing lineup with a perspective quite different from those who interpret these moves as commercial failure (like Nirvanix), customer abandonment (at worst) or perhaps bait-and-switch. Like Ben Kepes in his post on IBM, I note and am more heartily in alignment with Holger Mueller on the SCE announcements.

IBM is now selectively phasing out their initial offers (offered up under the SmartCloud Enterprise moniker), to be replaced by a 'composed' offering that relies on open cloud infrastructure (both CloudStack and OpenStack) delivered by Softlayer; storage solutions that are in lockstep with the maturity of OpenStack object storage as well as older forms more familiar to those with private SAN or NAS appliances; platform services built on Pivotal's Cloud Foundry and a stated reliance on BOSH as a general purpose utility for deployment of complex, composed applications; and… the list goes on.  

IBM's cloud thought leaders are incredibly active in their support of and contribution to open source cloud.  And, it's not just a marketing and sloganeering effort.  There is a real presence by the IBM cloud technology strategists like Angel Diaz at the Cloud Foundry Conference or Mac Devine's Keynote address recently the LinuxCon + CloudOpen conference. (Check here for the post-keynote panel discussion.)

Disparagement and Derision about Marketing and Packaging.  

As I started to read Enderle's article, I wondered at the likely reaction from the Clouderati who are now mostly engaged in word-wars about mindshare between AWS and OpenStack (with Open CloudStack making a guest appearance), the PaaS competitions which have emerged, and the various flavors of "software defined everything."  I tweeted the link and predicted that the Cloud commentary would be generally negative, derisive reaction.

I'm now going to ask for Clouderati restraint.  (… and if any of you comment about the oxymoronic nature of that concept, bear with me.)  Before throwing brickbats at IBM or at one another, take a moment to consider:

  • If this article were about ANY other major systems or services provider, would you have made the same comments?
  • In the face of a company that has signed up to embrace open source cloud and enhance it with specific added value options for their EXISTING as well as new customers, would you not have lauded the transition as being true to the open source software credo?
  • When you consider companies that have dominated their sectors of the technology market in the recent past - including IBM, Microsoft, Oracle, BEA, Red Hat - to what degree HAS marketing, positioning and attendance to the purchasing customer and the end-user been a major pillar of their strategy and ultimate commercial success?

Now re-read the Enderle article, and drop me a line (or a tweet at @rhm2k).  



A Decision Tier for Data Capture Networks

I'm a big fan of Adron Hall and understand the nature of this piece. But, I respectfully disagree, at least to the initial premise.

So I’ve been in more than a few conversations about data structures, various academic conversations and other notions about where and how data should be stored. I’ve been on projects and managed projects that involve teams of people determining how to manage data so that other people can just not manage data. They want to focus on business use and not the data mechanisms underneath. The root of everything around databases really boils down to a single thing – how can we store X and retrieve X – nobody actually trying to get business done or change the world is going to dig into the data storage mechanisms if they don’t have to. To summarize,

nobody actually gives a shit…

At least nobody does until the database breaks, or somebody has to be hired to manage or tune queries or something or some other problem comes up.

The point at which I take exception is when considering data collection and distributed data processing in conjunction with the Internet of Things (IoT) and, particularly, Industrial Internet.  Frankly, I have not yet encountered any group (commercial, open source, ...) to whom I can attribute a complete grasp of the IoT topologies required for efficient (intelligent) data acquisition and its retrieval or analysis.

That said, I fully agree that, once understood, the questions of where best to place data and the rise of the Intelligent Data System is our objective.

What would happen if the systems storing the data knew where to put things? What would be the case for providing an intelligent indexing policy or architecture at the schema design decision layer, the area where a person usually must intervene? Could it be done?

A decision tier that scans and makes decisions on the data to revamp the way it is stored against a key value, geo, time series or other method. Could it be done in real time? Would it have to go through some type of processing system? The options around implementing something like this are numerous, but this just leaves a lot of space for providing value add around the data to reduce the complexity of this decision making.

Thanks, Adron, for mooting the notion of a 'decision tier'.  I'm planning on using this mercilessly in my contemplations of Data Capture Networks (DCNs) for Industrial Internet.



How We Killed Privacy Long Before PRISM

Foreign Policy has published an excellent review of US privacy and surveillance law and its evolution (if one can call it that) during an age of unprecedented change in the use of personal and mass media, and technology change at speeds which make legislative response look positively glacial.

The idea that it has suddenly and suprisingly been revealed that we have little or no privacy with the NSA revelations is, quite frankly, bogus.  It's a theme generated by an over-heated journalism (looking for eyeballs) and the unfortunate over-use of privacy threat by those organizations which legitimately are pushing back on abuse, have been for years, and now see an effective way of getting it in front of the general populace.  I find both disturbing and hypocritical


How We Killed Privacy -- in 4 Easy Steps - By Daveed Gartenstein-Ross and Kelsey D. Atherton | Foreign Policy:

Privacy in 2013 does not exist as we knew it in 2000.
But don't be fooled: The almost complete erosion of what we would have considered our private spaces at the beginning of this millennium is not entirely -- nor even mainly -- a result of the National Security Agency's surveillance. While nobody should doubt that the government's electronic spying is intrusive, we largely let online privacy slip away without any assistance from security agencies. Each step along the way was, for the most part, understandable and reasonable rather than nefarious. But the fact is that privacy in the United States is not what it used to be, and until we realize that, our debateSave & Close about electronic privacy -- Manichean as it is, and focused almost exclusively on the relationship between the government and its citizens -- will fail to resurrect its value.
Four distinct factors have interacted to kill electronic privacy: a legal framework that has remained largely static since the 1970s, significant changes in our use of rapidly evolving technology, commercial providers' increasingly intrusive tracking of our every online habit, and a growth in non-state threats that has made governments the world over obsess about uncovering these dangers. Only by understanding the interaction between these factors can we begin the necessary discussion about what privacy means in the 21st century -- and how to forge a new social compact to address the issue.



Growth Factors: IaaS Numbers and Expectations.

Lots of new IaaS reporting going on this week.  Gartner has a new Cloud Magic Quadrant, and Synergy Research Group has a new report on cloud market size and growth.  

The problem I have with the IaaS revenue numbers comes down to this: AWS hides them. IBM has been called on the carpet for being 'over-inclusive' in its reporting. The combination of IaaS/PaaS numbers into one figure for Microsoft and Google disguises the fact that MSFT and GOOG both started in Platform Services first, and have only recently offered up 'pure' infrastructure services. 

The important factors of IaaS/PaaS accelerated growth are these:

[1] Infrastructure Service adoption -- new customers and the expanded use by existing customers -- in all market segments and among all the enterpise size classes is explosive.  The opinion that "enterprise won't put their [data | processing] in the Cloud" is bogus.  We're all -- ALL -- using cloud infrastructures, and will continue to add new data and processing to the cloud as comfort levels increase.

[2] The growth in IaaS and PaaS will continue on this path, and probably see acceleration, as enterprise and government customers become 'comfortable' with the upcoming offerings for cloud data management (e.g., security, privacy, distribution, meta-data generation, …) and for 'cloud automation' that includes APPLICATIONs, not just infrastructure configuration management. (We really need the cloud punditry to get straight on what it refers to when using the term 'orchestration.')

[3] Finally, just as Infrastructure Services understand the value of giving development and operational IT communities self-service and the ability to 'have it their way', they will recognize the need for fine-grained, customized alternatives for consumption and economic terms.  That is, more choice in the packaging and economic terms on which Infrastructure Services are procured.

On this last point: I expect considerable push back from enterprise customers on the measly number of choices they have to procure on-demand IaaS at either (a) rack rates using a credit card or (b) 'reserved' use in one-year or three year packages.  That is  far too little operational and economic choice.  The same kind of late-binding, fine-grained control the characterizes cloud development, deployment and functionality needs to be available to the businsess functions of IaaS/PaaS procurement, metering and billing.

More proof that Amazon still leads the IaaS pack, but watch out for those other dogs — Tech News and Analysis:

Amazon leads Google, IBM, and Microsoft in cloud. So what else is new? The fact that the other guys are growing like weeds.