Once upon a time there was DevOps...

Written by Damien Dubuc, on 06 February 2018

Today, we are living in a new era. Our new consumption patterns are based on multi-services and immediacy. The widespread use of applications via our smartphones is a perfect example. We want simple, functional, fast applications with frequent updates to quickly meet our new needs. Otherwise, we turn to the competition.

These new rules of B2C apply to B2B as well. Reducing Time-To-Market while ensuring the reliability of the application (or services) in production is the main challenge for companies in their digital transformation to retain their current customers and attract others through new competitive services available on the market.

Digital transformation impacts all sectors. IT is at the heart of these various sectors and has become an essential lever. IT must transform to help the business transform. Not all companies have been designed with an IT system based on principles of agility, automation, innovation... (Not everyone is Google, Amazon, Spotify, or Netflix). Companies have legacy systems to manage and a cultural, organizational, methodological, and IT history to evolve. DevOps allows transitioning from the old world to the new world of automation and even robotization.

DEV vs OPS? No, DevOps!

Here are some verbatims that will surely resonate with each of us:

  • [DEV] "There is an incident in production... It's not my problem."
  • [OPS] "There is an incident in production... The code must be bad."
  • "I wait several weeks to have my infrastructure."
  • "In development, my code works, so it must be the infrastructure that isn't tuned well."
  • "The DEV team doesn't know how the infrastructure works and misuses the features, so it's normal that it's slow."

These phrases make us smile, but they are still part of the cultural legacy of many companies. DevOps is primarily a culture whose objective is to break down the two silos of DEV and OPS ivory towers.

Whether on the DEV or OPS side, we are all IT and remember that a good IT recognizes itself as a "lazy" person, so it will prefer to automate all its repetitive and tedious tasks. This quality is a first step towards DevOps!

And at ANEO, how is DevOps implemented through Human DevOps?

During our support in transformations towards DevOps, we implement the following pillars:

  • A shared global culture between DEV and OPS in line with agile transformations and Software Craftsmanship (for both DEV and OPS, of course)
  • Automation to eliminate all unnecessary time losses from "human" processes (the software factory, the testing factory, continuous delivery, infrastructure as code...)
  • Shared Monitoring to better manage production, improve the application while avoiding expert debates on their own measures (DEV logs vs OPS infrastructure metrics)
  • Sharing between Business, DEV, and OPS for better collaboration and breaking down barriers based on tools, feedback for continuous improvement of the application and processes among the various stakeholders

With DevOps, one of the results is that a production deployment must become a non-event by offering an automated chain allowing for continuous delivery to continuous deployment.

ANEO supports its clients on three points:

  1. Cultural and organizational transformation to unite teams and orient them towards "services"
  2. Transformation of methods and tools (both DEV and OPS sides)
  3. Process automation

In addition to tools and methods, we want to start our approach by anchoring three main ideas among DEV and OPS:

  1. « All is software »
  2. « You build it, you run it »
  3. « Agile while stable »

« All is software »

The first idea is natural for DEV but less so for OPS. With the arrival of ITaaS (IT as a Service) in infrastructure teams, this approach revolutionizes the internal IT of large groups. ITaaS makes all infrastructure elements deployable and manageable by programming thanks to the open-source initiated by Amazon. Infrastructure experts must develop a service culture like Clouders (Azure, Amazon, Google...) who offer a significant marketplace of services and are real competitors. OPS mutation involves programming and software architecture practices. OPS must API-ize their infrastructure while securing it, making developers autonomous in setting up their own platform. This is the "self-service" infrastructure culture.

« You build it, you run it »

The second idea is to empower the developer to be responsible for the code delivered to production. Who is better able to support an application than the one who developed it? Thus, the developer who has solved some production incidents will be clearly aware of the stability of their application on the infrastructure concerned with the implementation of efficient monitoring.

« Agile while stable »

To speed up production while remaining stable, DEV and OPS must implement a Continuous Delivery chain or even Continuous Deployment (and tomorrow Continuous Operations). The mantra could be as follows: quickly delivered, quickly deployed, and quickly corrected or improved! Through automation, everything is coded, so the number of errors is limited or at least more easily detectable and correctable. Associated with the Monitoring of the DevOps culture, automation and the application must be continuously improved (code, testing, infrastructure deployment time, performance...).

A simple image to remember:

  • OPS must gain development skills to offer services that would be like LEGO bricks (IaaS) or entire walls of bricks (PaaS).
  • DEVs will be able to build their own structure with these different LEGO bricks or walls and thus provide a construction with stable and proven foundations and bricks.

The roles of DEV and OPS and their ecosystem will evolve. Some will see it as an opportunity, others as a threat or even anxiety. It is therefore important that this transformation be accompanied both in terms of cultural, organizational, and technological change. All this with the strong support of current managers who will have to rethink their role and posture by positioning themselves as leaders, vision carriers, and actors of the upcoming transformation.

To learn more about ANEO...

With its experience in organizational transformations, agile transformations, and technology, the consulting firm ANEO brings together a unique set of skills that allows us to support our clients in this transformation:

  • Expertise in transformation to understand the cultural and systemic challenges of redesigning an organizational and management model
  • Expertise in social sciences and change management to identify transformation levers and barriers to change
  • Expertise in IT to help you adopt DevOps in your organization
  • Concrete expertise in infrastructure and development to engage experts in an engaging dynamic
  • Expertise in innovative, digital, and participatory field training, helping to truly liberate the human capital of teams
  • Our own experience of transformation in a liberated company for more than 2 years with the effective implementation of agile principles, DevOps, and management 3.0.

To discuss your current challenges and the feedback from our clients whom we have supported both in cultural and organizational change, technical training, tooling, and extending agile transformation to the infrastructure team... come meet us at DevOps REX 2018, of which we are platinum sponsors!

Finally, like DevOps, which maintains a culture of Continuous Improvement, ANEO is in Continuous Hiring. So if you want to join our ANEO DevOps circle composed of consultants expert in both DEV and OPS technologies (because we must no longer make a distinction), as well as change management and coaching (technological - Infra as code, TDD, BDD, Container as a Service via Docker, OpenStack...) - and soft skills - Agile, Test & Learn, Communication...), do not hesitate to come see us and tell us about yourself!