Day #3 => DeployCon && Enterprise && Data Gravity

A lot of things were mentioned during the panels and sessions during DeployCon. Some things I agree with, some things I don’t.

The “Web Way”

There is one prevalent thing that came up over and over, the “web way”. What’s the web way? It is building horizontally, scalable, with RESTful APIs, and applications at an upper tier that are loosely coupled to any back end, and generally geographically independent (i.e. dispersed). This could be a mix of IaaS and PaaS, public or private, or a mix of all these things. Above all, it is about building horizontally instead of vertically, so that one can scale big, really big.

It was generally accepted in the dozens of conversations I had about the “web way” that the enterprise was only realizing that their idea of “big” and “redundant” just doesn’t even compare to what has been built using ideas from the Internet Generation. The Facebook PHP scale has blown away almost any enterprise’s scale problem, Twitter’s piping relationships across big data trumps most enterprises service bus concerns, Netflix’s streaming makes most enterprise’s issues with data transfer seem like a child’s game.

But all this isn’t to spite or hate on enterprise teams. They have an immensely important job to do. With all of these advances, and with management allowing enterprises to step closer into the community and get involved with things built the “web way” there is a massive upside. The enterprise is getting closer to cloud technologies. They’re learning how to leverage their on premise assets for various reasons, such as not throwing out the investment, and public or private cloud tech together.

Combing these things, and the complex needs of the enterprise together is all benefits, they’re getting the bleeding edge technology and research basically for free. All while effectively finding ways for a slow migration to not damage existing investments, day to day operations, and improve overall resiliency and services within the enterprise itself.

Overall, a huge win-win for everybody.

Data Gravity is Vital to Understand for Application Architecture

Dave McCrory kicked off the Data Gravity analogy to application and data source spacing. It is a pretty flawless analogy which brings us a much clearer understanding around application architecture when it comes to where data sources and where applications sit within a system. It also helps to define the value, cost, and gravity of data in relation to it’s location relative to the application.

With this topic brought up, every single tie it hinged around people fighting the gravity between data and applications. The difficulty in using cloud technologies, big data and other things is highly advantageous for companies, but as Dave points out in his presentations and writings about data gravity, it is getting exponentially more difficult to pull apart data and applications to rejoin them somewhere else. Such as the massively more scalable, powerful expansive public cloud.

These conversations lead to a growing opportunity space in the industry, moving data. This could be done with physical moves of drives, big pipes, pipes as a service or a host of other offerings. So far, very few are doing much in this realm, understandably considering the difficulty. But I’m betting that we’ll start seeing some serious efforts put into this. One of my personal notions is the idea of going cloud to cloud, we’ll see if anything pops up in the near future. If not, I might have to make a play on that myself.

…and last but not least, DeployCon!

DeployCon was the major reason I attended Cloud Expo. DeployCon was about PaaS Tech & the future movement of cloud technologies. Even though the rest of the conference, for most enterprises, seems bleeding edge, DeployCon was about the truly bleeding edge technology. Not only that, the elephant in the room is about the new king makers of IT, of technology, and of business in general; the software developers. Yup, I said it. We might want to stick IT on things here and there but the fact of the matter, over the next 10 years – if the IT moniker even sticks – it’s going to be more and more developers, business apps, and business development – not more network or system admin roles that are needed. The abstractions are pushing developers into an even more pivotal role in a company and providing even greater benefits to business. You can ask any number of people in the PaaS space, from AppFog, AppHarbor, Tier 3, Stackato/ActiveState all the way to Windows Azure (they need to actually show up next time, just saying!). They ALL SEE THIS HUGE CHANGE coming. Not only do they see it, they are experiencing the beginnings of the change.

So hats off to an excellent job Krishnan (@krishnan @ Rishidot Research) for putting this together! Wendy White (@wendywhite) for throwing in major logistics support and all the others too! I had a blast and am looking forward to the next event.

Thanks also to the fire starters and technologists on the panels! You guys know who you are, good job! Here’s a few parting shots of the awesome people I got to meet and talk tech with!

Parting Shots

@brianmmcclain & @markkropf enjoying a round of drinks!
@brianmmcclain & @markkropf enjoying a round of drinks! These guys are smart, seriously. (click for full size)
The one and only  @wattersjames the man with impeccable taste @mortonheroku and the man with the plan @guy_marion
The one and only @wattersjames (thanks for the fire starting!) the man with impeccable taste @mortonheroku and the man with the plan @guy_marion (click for full size image)
The beautifully witty @briellenikaido and @mortenheroku showing us the where the premo tastes are in New York City.
The beautifully witty @briellenikaido and @mortenheroku (GREAT Sake choice btw) showing us the where the premo tastes are in New York City. (click for full size image)

5 thoughts on “Day #3 => DeployCon && Enterprise && Data Gravity

  1. Does Facebook disappointing IPO accelerate the notion of going cloud to cloud in the enterprise?

    Enterprises deal in granular product or CRM-type data. Facebook deals in anonymizing data from a politically risky source, peoples identities they don’t own.

    Facebook becomes more valuable by hyper collecting identity data for better targeting adverts to maybes. Enterprises traditionally became more valuable building scale by acquiring businesses to improve information for profit and cost controls for certainty. Now however large enterprises are doing the exact opposite, divesting for flexibility, yet still need authentic granular information flows.

    The architecture to go cloud to cloud may be emerging in Federated SQL Azure sharding data ownership by hierarchical keys (@sqlservermIke). And in Cloudant (née CouchDB) play on replicating enterprise data (@mlmilleratmit)

Comments are closed.