Over a million developers have joined DZone.
{{announcement.body}}
{{announcement.title}}

Cloud Luminaries: Damon Edwards on Optimizing DevOps By Turning Everything Into A Service

DZone's Guide to

Cloud Luminaries: Damon Edwards on Optimizing DevOps By Turning Everything Into A Service

· Cloud Zone
Free Resource

Site24x7 - Full stack It Infrastructure Monitoring from the cloud. Sign up for free trial.

[This article was written by David Spark.]

Non DevOps environments are comprised of functional silos which can’t evolve as the business grows. They actually start to bog down the business, explained Damon Edwards (@damonedwards), Co-Founder and Managing Partner of DTO Solutions in our conversation at the 2014 Cloud Expo in Santa Clara, California.

What Edwards recommends is an effort to redraw the organization to eliminate the silos. Silos are the source of all these DevOps problems. Redraw and build teams around customer identifiable services.

“Anything that can’t go into one of these cross functional teams turn into internal service providers: security-as-a-service, QA-as-a-service, monitoring-as-a-service, environments-as-a-service, and so on,” suggested Edwards. “It’s very cloud inspired. It’s all about getting out of the way of the cross functional teams.”

For most businesses, a trouble ticket is a permission slip to do work, said Edwards. Instead, a trouble ticket should be just that. It should signal that there’s trouble that needs to be fixed. Everything else should be a series of self-service interfaces.

How do you get an existing organization into this mode? A reorganization isn't going to work because culture and process isn’t going to allow this to happen, said Edwards in his presentation “Without Self-Service Operations the Cloud Becomes Expensive Hosting 2.0.” Here's what Edwards recommends:

  • Turn information flow into artifact flow.
  • Insert verification points to drive feedback loops. No documentation for testing, it should simply be code that does this.
  • Drive all changes through a Software Delivery Life Cycle (SDLC).
  • For concerns that cross all teams, create services as mentioned above.
  • Start working like an internal service provider. Define your offerings. Tame the tool sprawl. Set-up self-service interfaces and secure access.

Only let trouble tickets be exceptions, not the daily work, said Edwards, because manual request queues lead to bottlenecks, increased lead times, reinforcement of organizational silos, and misinterpretation or omissions.

Site24x7 - Full stack It Infrastructure Monitoring from the cloud. Sign up for free trial.

Topics:

Published at DZone with permission of John Wetherill, DZone MVB. See the original article here.

Opinions expressed by DZone contributors are their own.

{{ parent.title || parent.header.title}}

{{ parent.tldr }}

{{ parent.urlSource.name }}