Features

Top Pressing Concerns for CI/CD in 2020

The tech landscape is changing rapidly. With the need for increased development velocity, CI/CD has become ubiquitous among most organizations. Yet, with the growing number of microservices, CI/CD may be tough to wrangle without the right strategy.

At the same time, it’s difficult to keep pace with the constant evolution around CI/CD. So, what is the state of the CI/CD space in 2020, especially in these turbulent times? What do stakeholders see as top pressing concerns companies should consider to advance their release fluidity?

For this article, I interviewed two stakeholders on the latest trends and industry cooperation around CI/CD. Read on to discover what two CI/CD thinkers, Tracy Miranda, board chair, Continuous Delivery Foundation (CDF), and Lachlan Donald, CEO, Buildkite, see are pressing concerns for CI/CD in 2020.

More Microservices, More Problems

With the number of microservices rising to the hundreds (or thousands) for some enterprises, maintaining consistent release strategies across the board could easily become a nightmare. If the goal is to replicate a unique CI/CD pipeline per microservice, pipeline automation and configuration management become top concerns.

“Automation will be key,” said Miranda. “We can no longer afford to be writing each CI/CD pipeline by hand; we will need new approaches.” She added that recent industry discussion shows “configuration management is a top concern for CI/CD practitioners.”

Tools like Jenkin X could enable this sort of automation, she noted. Companies should consider evolving their CI/CD pipelines with better configuration management.

Platforms Must Evolve

The recent “State of the Cloud Report” from Flexera found 39% of IT leaders saw implementing CI/CD in the cloud as a top priority. According to Miranda, the global crisis seems only to accelerate the need for cloud and CI/CD automation.

But, traditional CI/CD platforms aren’t always suited for modern development workflows with numerous pipelines. For Donald, an evolution in new tooling is absolutely necessary to help companies keep pace.

“We’re at a point where organizations need to completely rethink their approach to CI/CD,” Donald said. “Traditional CI/CD platforms are not suited for modern development workflows.” He cites slow speed, unreliability, scalability limitations and security risks as common hangups among many CI/CD tools.

Lack of Clarity in CI/CD Market

The market has expanded significantly to address the gap in CI/CD tooling effectiveness. According to Miranda, this has led to a continuous deployment tooling ecosystem wrought with interoperability and competing styles.

“My biggest concern is the proliferation of tools and platforms with a lack of interoperability and lack of clarity of how to best use them or whether they are fit for purpose,” Miranda said.

These differences make it difficult for developers to find the right integrations that work for their environments, leading to wasted time and effort. “Developers will experience growing pains with their CI/CD platforms as they work to keep up with the constant demand for quality software,” acknowledged Donald.

Lack of Standards

Part of the lack of clarity resides in the fact that there are little to no standards emerging in the CI/CD space. Aside from Accelerate rising as the “bible” for software delivery, Miranda sees a lack of standards in the CI/CD space.

“Not having standards or de facto standards is a problem,” she said. “For example, why is there no de facto standard for reporting pipeline test results?”

Advancing Open Source CI/CD Tools

To improve Interoperability throughout the CI/CD industry, bodies such as CDF are working on setting general guidelines. Miranda described how CDF is accomplishing this, by supporting a CDF Interoperability working group, sharing CI/CD practices and maintaining open source CI/CD projects. Two CDF projects she finds quite compelling are Tekton, a pipeline execution engine, and Jenkins X, a new CI/CD platform for pipeline automation.

Cloud vs. On-Premises: A False Dichotomy?

In analyzing the CI/CD tooling ecosystem, Donald recognizes a recurring dichotomy; that companies must choose between the security of on-premises and the speed of cloud-native. In reality, this dichotomy may be false.

“It is possible to have the best of both worlds,” he said. “A hybrid strategy [is] the solution for testing and deploying software at scale without compromising security or performance.’

As opposed to cloud tooling, self-hosted CI/CD control planes could grant developers more code control and help meet security standards. “I believe self-hosted agents are always a good idea,” Donald said.

As we recently covered, a hybrid cloud approach is picking up steam. Similarly, Jenkins X is designed to run on any Kubernetes cluster, regardless of on-premises or cloud, demonstrating industry movement toward tooling neutrality and hybrid motivations.

Moving Beyond PASS or FAIL

There is a reductive tendency in CI/CD to simply check whether a build passed or not. But, this mindset misses an enormous amount of useful information and undervalues the test and release suites themselves.

“Teams should get granular about what worked and what didn’t,” Donald said. DevOps teams must evolve beyond a “PASS or FAIL” mindset to analyze more minute aspects of the build process itself.

For Donald, this also means shifting to treat test code performance “as a first-class citizen, applying modern development and scaling practices to it and allocating the necessary resources to it.”

Future Evolution: Progressive Delivery and Hybrid Approach

In closing, I asked Miranda and Donald their thoughts on the future CI/CD evolution. They shared their perspectives on where CI/CD needs to evolve.

Miranda views progressive delivery as the next milestone. With progressive delivery, teams deploy services in a staged manner, she said. “Tools, technologies and processes that fall under the ‘progressive delivery’ umbrella will be the next milestones for improving deployment speed and efficiency.”  Building blocks such as services meshes could help advance this process, she noted.

Donald stressed the need for a hybrid approach to CI/CD. A neutral control plane could grant better control and future-proof continual release strategies.

Assessing the CI/CD Landscape

In summary, our panel shared thoughts on the top concerns for CI/CD in 2020:

  • Platforms must evolve: Teams should rethink their approach and emphasize automation in pipelines.
  • Lack of clarity: There are few standards and a lack of clarity among tools. This confuses developer adoption.
  • Interoperability: A focus on interoperability is critical toward advancing the industry.
  • Unhelpful dichotomies: The cloud-native versus on-premises divide may not be helpful for a hybrid strategy; thus, it’s recommended to seek out platform-neutral tools.
  • Move beyond PASS or FAIL: Switch the mindset to “Test code = application code.”

High performant organizations must scale their technology tremendously. With this desire comes a mounting strain on how companies adopt a continuous deployment strategy. Without a proper CI/CD plan, business agility is sacrificed.

As an ending note, I absolutely love Donald’s statement: “A good litmus test for how well a company is run is their approach to CI/CD.” Perhaps what all can agree on is that companies must prioritize CI/CD, as it is now business-critical. Otherwise, deployments may be numbered for those organizations that don’t adapt.

Bill Doerrfeld

Bill Doerrfeld is a tech journalist and analyst. His beat is cloud technologies, specifically the web API economy. He began researching APIs as an Associate Editor at ProgrammableWeb, and since 2015 has been the Editor at Nordic APIs, a high impact blog on API strategy for providers. He loves discovering new trends, researching new technology, and writing on topics like DevOps, REST design, GraphQL, SaaS marketing, IoT, AI, and more. He also gets out into the world to speak occasionally.

Recent Posts

AIOps Success Requires Synthetic Internet Telemetry Data

The data used to train AI models needs to reflect the production environments where applications are deployed.

11 hours ago

Five Great DevOps Jobs Opportunities

Looking for a DevOps job? Look at these openings at NBC Universal, BAE, UBS, and other companies with three-letter abbreviations.

20 hours ago

Tricentis Taps Generative AI to Automate Application Testing

Tricentis is adding AI assistants to make it simpler for DevOps teams to create tests.

3 days ago

Valkey is Rapidly Overtaking Redis

Redis is taking it in the chops, as both maintainers and customers move to the Valkey Redis fork.

3 days ago

GitLab Adds AI Chat Interface to Increase DevOps Productivity

GitLab Duo Chat is a natural language interface which helps generate code, create tests and access code summarizations.

4 days ago

The Role of AI in Securing Software and Data Supply Chains

Expect attacks on the open source software supply chain to accelerate, with attackers automating attacks in common open source software…

4 days ago