Cloud // Platform as a Service
Commentary
3/24/2014
10:55 AM
Charles Babcock
Charles Babcock
Commentary
Connect Directly
Twitter
RSS
E-Mail
50%
50%

Agile, DevOps, Cloud: IT's New Trinity

Agile, DevOps, and cloud form an almost-holy triumvirate with a common theme: Software rules.

As software has grown in importance, overshadowing the once-riveting details of chips and hardware, a new language has crept into computing's vernacular. Where once we talked in silicon yields and gigahertz speeds, we now speak of scrums and sprints, DevOps and Continuous Delivery, and most of all, workload deployments to the elastic cloud's infrastructure.

The genie in the bottle is no longer the secrets in the hardware; it's the invisible functions of the software.

Even the terminology of software has changed. In the past, we might have declared new development had nothing to do with maintenance of the datacenter's legacy systems. Development was novel and exciting, a thing apart. Nor did development methodologies have anything to do with production deployments. And certainly an enterprise software project could proceed and have nothing to do with the cloud.

But nearly everyone now knows these old verities are less likely to be true. Most of us would say all these things are now related, and some would say, closely related. I would go a step further. All of these things are the same idea, expressed in different ways. They reflect the importance of the leading ingredient of business survival: new software.

[Want to learn more about agile development working with DevOps? See Where Agile Development Fails: IT Operations.]

The world runs on software. We have software-defined virtual servers, software-defined networks, and, soon, software-defined datacenters. It sometimes seems the world is so dependent on software that we need it to make the sun rise and set on time. Software-defined sunsets. Thank goodness, that hasn't happened yet.

Although agile development is all about software, it's not really a set of technology goals. At its heart, it's really a belief system. Among other things, its tenets are: Software must reflect the needs of the business; it must be developed rapidly and timed to the needs of the business; it must be delivered already tested and ready to run. In short, the software must better reflect the challenges and realities with which the business is dealing. The agile oath is: We'll do this better than the waterfall method did.

Likewise, DevOps, though seemingly a set of technology methodologies and disciplines, is more like yet another belief system. Rackspace CTO John Engates, in a column for InformationWeek, noted how DevOps "holds certain truths to be self-evident." They are: "Shipping code faster and more error free is inherently good. Automated testing at scale makes for a better, more secure product. The real value of engineering talent is not the manual orchestration of tasks, but the insight and creativity to solve interesting, real-world problems." In other words, the highest calling of an engineer is in getting the software to reflect the realities of the business and respond with an answer to the challenge of the world as we find it. It's another expression of how survival depends on software.

So how does cloud computing match up with agile and DevOps? Is cloud a belief system, too? That doesn't seem too likely. Gotten any feelings of awe and omnipotence out of Amazon lately? Probably not, unless you're looking at your bill.

But still, think about it. Once software has captured the business's response to its challenges, it needs someplace to run. That used to be -- often still is -- the enterprise datacenter. But what if the engineers have done such a good job, solved the challenge so well, that new customers are arriving in droves? The datacenter server cluster is chugging along as best it can, but no one anticipated this much success when the IT budget was drawn up.

Michael Jackson was a rock star who, through much of his life, created and sold a steady amount of music. When he died, there was a sudden and seemingly insatiable appetite for his music. The online music store servers chugged and chugged, but they couldn't come anywhere near to keeping up with demand. The wave crested, broke, and receded. So did the opportunity.

To many in this New Age of Software, it's an article of faith that software that reflects big ambitions must have a big platform on which to run. It may not need it all the time, but when its moment of truth arrives, the platform needs to be as big as the opportunity. What agile and continuously improved software needs is an elastic resource, one that can expand with the business. You can't draw a line where that expansion might stop. It needs, theoretically, to be infinite.

That's not the cloud, but almost. The cloud offers its users the illusion of infinity. Somewhere there's a physical limit, but with many thousands of users and some overhead margin for each, there's a very large surplus capacity built into the platform that can be tapped at any one time. And if it gets tapped out, there may be an associated cloud datacenter with its own surplus. To the user who needs that surplus, the cloud is an infinite resource.

It's not enough to compose excellent software that constantly adjusts to the changing nature of the business. That software must have a suitable place to run, one whose merits can reflect its ability to seize the moment and capitalize on it. That doesn't mean that we're searching anymore for the latest chips, servers, and switches. It now means the software cloud -- which can fire up virtual servers, networks, and storage; compensate for any hardware component's failure; and keep all systems running. Finally, we have the intelligent environment we've been seeking to run our brilliant business software.

Agile, DevOps, and cloud are an almost-holy triumvirate in terms of the belief systems they marshal and engineering aspirations they contain. They're all trying to do the same thing: give software, and software alone, its chance to triumph over the many competitive hurdles and perils that await us. If software is everything, then embracing the mantras of agile, DevOps, and cloud is one way to invoke its power. And in doing so, we hope to move ourselves, our organizations, and our whole culture another step closer to Nirvana.

Engage with Oracle president Mark Hurd, NFL CIO Michelle McKenna-Doyle, General Motors CIO Randy Mott, Box founder Aaron Levie, UPMC CIO Dan Drawbaugh, GE Power CIO Jim Fowler, and other leaders of the Digital Business movement at the InformationWeek Conference and Elite 100 Awards Ceremony, to be held in conjunction with Interop in Las Vegas, March 31 to April 1, 2014. See the full agenda here.

Charles Babcock is an editor-at-large for InformationWeek, having joined the publication in 2003. He is the former editor-in-chief of Digital News, former software editor of Computerworld and former technology editor of Interactive Week. He is a graduate of Syracuse ... View Full Bio

Comment  | 
Print  | 
More Insights
Comments
Newest First  |  Oldest First  |  Threaded View
Charlie Babcock
50%
50%
Charlie Babcock,
User Rank: Author
3/27/2014 | 6:04:12 PM
Past may be prologue to but doesn't pre-determine the future
Steve, you have a point. The primary purpose today for agile is to speed up modification of legacy systems. But there's no reason its principles won't be applied to writing of next generation apps and in fact they already are. The past doesn't always predetermine the future. Dev teams will comprehend the cloud and want to deploy to it.
SteveStrutt
50%
50%
SteveStrutt,
User Rank: Apprentice
3/27/2014 | 1:19:41 PM
Re: Sophistircated software running on cheap hardware
I differ. The value of Agile and DevOps is to realise business value quickly and monitise opportunities. In this respect they are is ideally married with cloud to deliver business outcomes. The economic cost of infrastructure is secondary as long as the cloud platform is agile and scalable. 

I see the value of working with clients to bring DevOps and agile to existing applications and function, is in enabling change and progression in previously static legacy systems that have been holding businesses back.

The economic argument for use of commodity 2 9's hardware brings on a different challenge of application design and development. How many developers have expertise in developing resilient 5 9's apps on 2 9's hardware and can use frameworks like Netflix OSS. DevOps and Agile are good for iterating application function based on user feedback, poor at interating non-functional characteristics as feedback is over a much longer time frame. Which Netflix found to their cost a few Christmas's ago.

Core banking and blue light services are unlikely candidates, long with many others. Though those systems certainly benefit from agile and DevOps.
dbartoletti570
50%
50%
dbartoletti570,
User Rank: Apprentice
3/26/2014 | 12:55:03 PM
Couldn't have said it better myself
Excellent article, Charlie! Great minds :-)
Charlie Babcock
100%
0%
Charlie Babcock,
User Rank: Author
3/25/2014 | 6:40:25 PM
Sophistircated software running on cheap hardware
Good summary, J.P. Morganthal. Randy Bias strikes again. He's said many memorable things. But I'd add it's not just the resilient software on cheap hardware. It's also the simplified, standardized environment and the ability to apply automated processes to it. But at the core, the economic argument, sophisticated software running on cheap hardware, stands.
jpmorgenthal-tw
50%
50%
jpmorgenthal-tw,
User Rank: Apprentice
3/25/2014 | 9:57:19 AM
Is it about delivering at scale
For a long time I struggled with why I should recommend a customer pay to migrate a steady-state workload to any type of cloud environment (private, public, hybrid). After all, it doesn't see the spikes mentioned in this article, there's a nominal % growth year over year, the operations team have a complete Runbook for operating the system through any type of disaster. There's so many other areas where IT could more effectively spend their money than to take this workload and migrate it to a cloud environment.

What convinced me of the importance of taking this step was listening to Randy Bias at a recent DevOps Meetup in Chicago. I've paraphrased it so much, I don't remember his original statement, but for me it works out to there's a major fundamental benefit from delivering 5 9's software running on 2 9's hardware versus building 5 9's in hardware.

And there it is, the gold at the end of the rainbow for DevOps.

The mission that could only effectively be met by development & operations working hand in hand as an effective team with appropriate controls and communications. The value proposition is significant. If you don't believe, read the stories coming out of Google, Facebook & Twitter. Resiliant software designed to operate in the face of hardware failure with no/minimal loss of data running on commodity hardware that can be continually be expanded without requiring maintenace windows or reconfiguration, operates across geographic distances so that loss of an entire facility constitutes only a minor disruption in service all done at a fraction of what it would cost to operate using traditional enterprise hardware alternatives.
cobiacomm01
50%
50%
cobiacomm01,
User Rank: Apprentice
3/24/2014 | 6:12:27 PM
DevOps meets Agile ALM in the Cloud
Great post Charlie!    Agile ALM accelerates software development iterations, DevOps accelerates deployment, and Cloud accelerates execution.      How many environments and tools unify the trinity points?     Have you taken a look at WSO2 App Factory, a DevOps PaaS?
Charlie Babcock
50%
50%
Charlie Babcock,
User Rank: Author
3/24/2014 | 12:54:54 PM
Software is taking over the world
That's about right, Lorna. LIke the diner analogy. Software is taking over the world, and if you don't have the most efficient platform on which to run it, you'll find all kinds of disadvantages slowly accruing to your account. Sometimes that platform is on premises, sometimes in the cloud. What we really want is close to 100% utlization of the resource in either place.
Lorna Garey
100%
0%
Lorna Garey,
User Rank: Author
3/24/2014 | 11:25:27 AM
It's about the spikes
Charlie, thought about this yesterday when I passed a local diner. The place is usually 65% full, except on Sunday between 10:00 and noon, when the line is out the door and plenty of potential customers decide not to wait and drive on by. Every week, the owners lose money because they can't handle a predictable spike in demand. Companies that take to heart the concept of Agile and DevOps -- which quite possibly will lead to a demand spike -- but don't add that cloud angle risk leaving millions on the table.
Google in the Enterprise Survey
Google in the Enterprise Survey
There's no doubt Google has made headway into businesses: Just 28 percent discourage or ban use of its productivity ­products, and 69 percent cite Google Apps' good or excellent ­mobility. But progress could still stall: 59 percent of nonusers ­distrust the security of Google's cloud. Its data privacy is an open question, and 37 percent worry about integration.
Register for InformationWeek Newsletters
White Papers
Current Issue
InformationWeek Tech Digest, Nov. 10, 2014
Just 30% of respondents to our new survey say their companies are very or extremely effective at identifying critical data and analyzing it to make decisions, down from 42% in 2013. What gives?
Video
Slideshows
Twitter Feed
InformationWeek Radio
Archived InformationWeek Radio
Join us for a roundup of the top stories on InformationWeek.com for the week of November 16, 2014.
Sponsored Live Streaming Video
Everything You've Been Told About Mobility Is Wrong
Attend this video symposium with Sean Wisdom, Global Director of Mobility Solutions, and learn about how you can harness powerful new products to mobilize your business potential.