Last night some of you may have received an announcement from EMC discussing their Pivotal Initiative. Many of you reading this may not exactly be clear on what, how or why EMC is doing. Let me take this opportunity to clear this up.
Changing IT consumption models – what is is driving the Pivotal initiative
First things first, we are in the middle of a couple SIGNIFICANT transitions in our industry. Many people call this cloud computing, but it is much larger then that.
What we are truly in the middle in is a shift of the consumption model of IT resources. For the past ten to fifteen years consumption and spend has been centered in and around IT organizations, generally under the command and control of a CIO. These organizations would work with the business, and sometimes the applications teams to capture requirements and create infrastructure to service their internal customers needs.
Over the past ten year so so (really since SOX) the focus of many it organizations has shifted from Information Technology, to Infrastructure Technology. Significant amounts of time focus and budget (commonly measured at north of 60-70%) of most IT organizations are spent just keeping the lights on. This creates limited additional value to the companies that they are part of, as most importantly, is now a key limiter in application development and release cycles (the true lifeblood of many companies – their ACTUAL product).
The rise in power and maturity of software development
This slow old IT worked in the past when software development life cycles were also old and slow. Classic SDLC based development shops would sometimes take years to release code. In this world, slow and bloated infrastructure operations organizations (classic IT) was not the roadblock. Everything was fine.
Over the past few years however, things like Agile Software Development has taken the world by storm. In this world, development teams release code weekly, if not daily. This code is of higher quality, and more feature rich then ever before.
What this has caused, is an imbalance between the needs of software development organizations and the capabilities of the IT organizations that support them. To meet that need, many development organizations looked to companies that could provide “Agile infrastructure”. Amazon Web Services (AWS) is one of these companies, and has seen extreme growth because they can serve this need.
Cloud Computing is the first shift – DevOps / NoOps is the next
Cloud computing, defined as the programatic, fractional consumption of elastic services (Amazon Web Services as ONE of the providers of this) is the first mega growth area that has benefited from this development shift.
It is however still not super simple for developer to write code for Amazon, or free OpenSource Amazon “Clones” like Eucalyptus and OpenStack. They have to have a level of knowledge of how to consume, manage, monitor and scale the services offered from these providers.
Remember, the goal of any development organization is to abstract complexity from the underlying layers of the App. IaaS tools and platforms with good web services interfaces are only the first step in making this happen.
Enter the PaaS stacks – the cloud for cloud
PaaS (Platform as a Service) is a layer that runs on top of cloud platforms. Generally PaaS stacks abstract all the unique features of any cloud platform, allowing a developer just just focus on writing good functional code. The PaaS stack then handles connections to database services, autoscaling, etc (example is CloudFoundry and BOSH).
This abstracts the cloud provider, and sets the foundation for true cloud mobility. This I believe is the future, and will be the path to large scale cloud adoption.
Where does Greenplum and Analytics fit into this?
Data Science is the next great application consumer. Organizations that employ data science as part of their business process create and consume EMENSE amounts of data. Needless to say, companies like EMC love this. You do eventually have to put that data somewhere…
What does EMC / VMware’s Pivotal initiative mean in this context
Open software and PaaS stacks are clearly the future consumption target for the evolved development team. The challenge is that this is because Open Source software is also free software (at least partly). Big publicly traded companies have a real problem executing on both developing, and integrating free software into their sales models.
In some companies, like EMC. I hear that it actually is impossible to release Open Source code written in certain business units. The problem with this is, that releasing certain software into the wild for free actually drives consumption of big expensive blinking enterprise hardware.
Creating a new company, filled with talent focused on PaaS and Open Source is Brilliant
1,400 employees are supposed to join this new company (Pivotal for now), focused on PaaS, Analytics at scale and Open Source. This positions Pivotal (or whatever it will be called) to be a dominant player in this shift in consumption models. It allows EMC to stay relevant as classic storage becomes passé and compute can in any location, ALSO including your data center.
Remember when EMC bought VMware
Nobody understood what was going on when EMC bought VMware. What they were really doing was ensuring that a new consumption model that would grow demand for their product sets would thrive. It also creating a landing spot for some of the brightest minds in the industry (Remember, Chad and the vSpecialist teams came out of there). Pivotal initiative provides the same play in the “new world” of IT.
What does this mean to you
This industry is changing, at an amazing pace. The classic world of IT is evaporating, and forming into something amazingly interesting and relevant. EMC has gone all in now, and committed and amazing number of resources to making true cloud application consumption “normal”. This will provide with each and every one of us new tools and techniques to use, and also learn.
With all shifts, some people and teams will rise, some will get left behind. My advice, pick up the old programming books, and start to think like a developer… because that is where our world is going.
- My name is Colin, and I support OpenStack
- VMware’s Acquisition of Nicira – VMware confirming the hypervisor is dead
- Cisco announces it’s own OpenStack Distribution – What will the effect be on VMware?
- VMworld 2009 Schedule
- Cisco releases Nexus 1000V virtual switch for VMware
- Cisco EMC and VMware partneship VCE VBlocks Acadia and the Partner Ecosystem