The superorganism of the Continuous Delivery pipeline

Three ant behaviors every DevOps team needs.

Here’s an interesting fact: there are about 10 quadrillion ants on the planet, comprising up to 25 percent of the earth’s animal biomass. That’s a scary population of ants. But, thankfully, they are not bringing mass chaos to the planet because their colonies operate as well-orchestrated superorganisms—pooling resources to accomplish things for the greater good of the colony. And the cool thing is that they make better decisions and fewer mistakes operating together.

Check out the three ant behaviors any good DevOps team should acquire.

Ant behavior #1: Specialized roles working seamlessly

Both an ant colony and the continuous delivery pipeline can be thought of as a factory. Like in a factory, there is a certain level of specialization for the different tasks that must be accomplished. In the ant world, you’ll find the queen who lays the eggs and the worker ants who take care of the nest, gather food or defend the nest. In the software factory, this might be labor managed by a specific team or a specific tool, from planning into production.

But the interesting thing about ants is that there’s a certain level of autonomy. The ants can act on their own, finding new food sources or avoiding obstacles, but they always work toward the greater good. Through communications ranging from physical touch to pheromones to vibration, they ensure that they stay connected and share their knowledge—like marking a trail, sending out warnings or highlighting a new food source.

This is where the continuous delivery pipeline can fall down. How can you help connect the dots between many teams, tools and sometimes conflicting priorities to ensure the best outcomes?

For example, testing is one of the biggest obstacles to fast, efficient application delivery, often treated as a specialized, isolated workflow operating with traditional, manual testing procedures and technologies.

You must eliminate any silos—essentially creating a seamless team including developers, QA, release managers, operations and more—which enables you to dramatically improve how long it takes to get a code change into production. To achieve this, you need:

  • An automated and orchestrated DevOps toolchain end-to-end
  • Automated feedback loops and communications
  • End-to-end tracking and visibility of a change from planning into production

A single control point for the DevOps toolchain can be like the pheromones of the ant world, connecting everyone together and driving action.

Ant behavior #2: Smarter decisions together

When you combine the many tiny brains of an ant colony, you get a smart superorganism. For example, many individual ants may go out to explore new food sources, and, upon their return, the collective knowledge is processed by the colony to determine the best location. As a collective, the ants can weigh many variables and make fewer mistakes. This is distributed intelligence.

Now, I’m not saying DevOps teams are filled with tiny brains! But the brains often only see one side of the bigger picture. It’s important in a modern software factory to enable that distributed intelligence, pulling together the knowledge from all areas of the end-to-end application lifecycle to enable faster, more frequent releases and better customer experiences.

For example, you can improve quality by including testing phases and tools in an automated release workflow. By orchestrating the testing nodes, you can perform testing in parallel, accelerate development fixes with faster feedback and enable real-time, informed decisions based on release health.

Ultimately, the more collective intelligence you build into the release process, the better your apps and your processes become. Data derived from a connected toolchain can enable real-time actions (like identifying and eliminating bottlenecks) or improving processes. It enables you to look beyond one release and see how your entire pipeline is operating.

Ant behavior #3: Adaptive and dynamic

Because ants are so in tune with each other, their ability to adapt to their world is outstanding. When a food source dries up, the pheromone path to the food starts to dissipate, leading to new foraging. Some ants live is spatially distinct colonies but stay socially connected. This helps them mitigate risks to the colony and leverage their resources more effectively. Ants can even dynamically switch from one task to another based on cues from the environment and from other ants. For a modern software factory, this means the system must be connected and open for change.

The DevOps toolchain will be constantly evolving as new tools emerge that solve new problems or do a better job solving existing problems. The mechanisms to orchestrate the toolchain can’t be rigid or prone to excessive maintenance. They must be able to scale and manage the scope of true continuous delivery.

DevOps roles also are evolving where everyone is responsible for quality. Testing is shifting left, with the mantra of fail fast, get feedback, learn and improve. Release teams are juggling daily and weekly small batch releases versus monolithic quarterly releases.

This requires better planning, complete visibility for everyone, instant communications, automatic promotion through the pipeline, and analytics for sourcing real-time and longer-term improvements.

Help your continuous delivery colony thrive

CA Continuous Delivery Director orchestrates continuous “everything”—development, testing, release and improvement—so that you can continuously deliver high quality applications to your customers. It is DevOps tool agnostic, meaning that you can connect your open-source, commercial or home-grown tools for full pipeline visibility and streamlined operation.

With CA Continuous Delivery Director, you can plan, schedule, govern and monitor multi-app, concurrent releases across all phases of the application lifecycle, handling dependencies and conflicts. You also can orchestrate your testing workflow—automating the promotion of builds, feedback loops and risk assessment. And most importantly, you get actionable analytics for identifying bottlenecks, making smarter go/no go decisions and improving processes, releases and apps.

Start adopting these ant behaviors today by trying the new SaaS release of CA Continuous Delivery Director for free.


Patricia Johnson is a product marketing manager. She has been in the high-tech industry for…

Comments

rewrite

Insights from the app driven world
Subscribe Now >
RECOMMENDED
The Sociology of Software >How (Not) to Lie with Data Visualization >DevOps and Cloud Computing: Exploiting the Synergy for Business Advantage >