Azure DevOps vs. CircleCI vs. GoCD

Azure DevOps works well when you’ve got larger delivery efforts with multiple teams and a lot of moving parts, and you need one place to plan work, track it properly, and see how everything links together. It’s especially useful when delivery and development are closely tied and you want backlog items, code and releases connected rather than spread across tools. Where it’s less of a fit is for small teams or simple pieces of work, as it can feel like more setup and process than you really need, and non-technical users often struggle with the interface. It also isn’t great if you want instant, easy programme-level views or a very visual planning experience without putting time into configuration.

Incentivized

Based on our experience, CircleCI is well-suited for automating mobile app release cycles. For example, to release an iOS app, you would need to build, sign, and upload it to TestFlight, which requires a dedicated Mac in the office. But with CircleCI, you can have macOS executors, so you don't have to manage a physical build machine. Another benefit is that CircleCI's certified AWS Orbs abstract away complex authentication and deployment logic, allowing us to build, push, and deploy Docker containers to Amazon ECS with minimal configuration and high reliability. CircleCI is less suited for smaller projects where the development and deployment are not that extensive, for example, a static site. Once you have built a static site, you probably won't make any further changes, so there's no point in paying for it.

Incentivized

Previously, our team used Jenkins. However, since it's a shared deployment resource we don't have admin access. We tried GoCD as it's open source and we really like. We set up our deployment pipeline to run whenever codes are merged to master, run the unit test and revert back if it doesn't pass. Once it's deployed to the staging environment, we can simply do 1-click to deploy the appropriate version to production. We use this to deploy to an on-prem server and also AWS. Some deployment pipelines use custom Powershell script for.Net application, some others use Bash script to execute the docker push and cloud formation template to build elastic beanstalk.

Incentivized

• Utilize Git as a repository to share work between multiple users
• Ability to configure Pipelines to build containers to run virtual deployments and testing scripts.
• Split individual tasks and relate to master documents for quick navigation and ability to see overall picture of project.
• Track status of each task
• Integrate with Git to utilize branches, merging, approvals, history, etc.

Incentivized

• Automated builds! This is really why you get CircleCI, to automate the build process. This makes building your application far more reliable and repeatable. It can also run tests and verify your application is working as expected.
• Simple. Unlike Jenkins, Teamcity, or other platforms, CircleCI doesn't need a lot of setup. It's completely hosted, so there's no infrastructure to set up. The config file does take a bit to understand, but if you follow their example and start with something small and add to it, you can get it up and going quicker than it first looks.
• Scales easily. Again, since it's all cloud-based, you don't have to manage or scale infrastructure. Simply subscribe to the number of containers you want, and scaling up just means buying more containers.

Incentivized

• Pipeline-as-Code works really well. All our pipelines are defined in yml files, which are checked into SCM.
• The ability to link multiple pipelines together is really cool. Later pipelines can declare a dependency to pick up the build artifacts of earlier ones.
• Agents definition is really great. We can define multiple different kinds of environments to best suit our diverse build systems.

Incentivized

• I did mention it has good visibility in terms of linking, but sometimes items do get lost, so if there was a better way to manage that, that would be great.
• The wiki is not the prettiest thing to look at, so it could have refinements there.
• It could improve the search slightly better.

• While configuration is easy, the config files can get very very long.
• Price compared to some alternatives that are cheaper / free. Especially so if you are running multiple containers in parallel.
• Have experienced numerous outages (3-5) in the last few months where CircleCI has been down.
• Web documentation and tutorials haven't been as good as some of the competitors.

Incentivized

• UI can be improved
• Location for settings can be re-arranged
• API for setting up pipeline

Incentivized

I don't think our organization will stray from using VSTS/TFS as we are now looking to upgrade to the 2012 version. Since our business is software development and we want to meet the requirements of CMMI to deliver consistent and high quality software, this SDLC management tool is here to stay. In addition, our company uses a lot of Microsoft products, such as Office 365, Asp.net, etc, and since VSTS/TFS has proved itself invaluable to our own processes and is within the Microsoft family of products, we will continue to use VSTS/TFS for a long, long time.

Incentivized

It's a great help to get more information about new feature release and stay updated on what the dev team is working on. I like how easy it is to just login and read through the work items. Each work item has basic details: Title, Description, Assigned to, State, Area (what it belongs to), and iteration (when it’s worked on). See image above.They move through different states (New → Discovery → Ready for Prod → etc.).

Incentivized

The reliability & speed, it just works. The ability to spin up macOS runners and Docker containers on demand without managing hardware is a huge win. The Orbs system makes integrating with AWS and Slack incredibly easy, saving us weeks of custom scripting and providing real-time updates in our Slack channel. This makes it easy for us to track and ensures that everyone involved knows the status. Of course, it has drawbacks related to configuration complexity and, in some cases, cost transparency, but overall, it is an industry-standard, robust tool that solves our core infrastructure problems well.

Incentivized

It's pretty snappy, even with using workflows with multiple steps and different docker images. I've seen builds take a long time if it's really involved, but from what I can tell, it's still at least on par if not faster than other build tools.

Incentivized

When we've had issues, both Microsoft support and the user community have been very responsive. DevOps has an active developer community and frankly, you can find most of your questions already asked and answered there. Microsoft also does a better job than most software vendors I've worked with creating detailed and frequently updated documentation.

Incentivized

Unless you have a reasonably large account, you're going to be mainly stuck reading their documentation. Which has improved somewhat over the years but is still extremely limited compared to a platform like Digital Ocean who invested in the documentation and a community to ensure it's kept up to date. If you can't find your answer there, you can be stuck.

Incentivized

Was not part of the process.

Incentivized

Microsoft Planner is used by project managers and IT service managers across our organization for task tracking and running their team meetings. Azure DevOps works better than Planner for software development teams but might possibly be too complex for non-software teams or more business-focused projects. We also use ServiceNow for IT service management and this tool provides better analysis and tracking of IT incidents, as Azure DevOps is more suited to development and project work for dev teams.

Incentivized

Jenkins is usually self-hosted, Travis CI's infrastructure is largely unreliable (lots of tests time out for no discernable reason), and Semaphore encourages you to configure your CI/CD from a web UI. We like CircleCI because its hosted, our tests run largely as expected on their infrastructure, and we can configure it from a config file that we track in GitHub.

Incentivized

GoCD is easier to setup, but harder to customize at runtime. There's no way to trigger a pipeline with custom parameters.

Jenkins is more flexible at runtime. You can define multiple user-provided parameters so when user needs to trigger a build, there's a form for him/her to input the parameters.

Incentivized

• We have saved a ton of time not calculating metrics by hand.
• We no longer spend time writing out cards during planning, it goes straight to the board.
• We no longer track separate documents to track overall department goals. We were able to create customized icons at the department level that lets us track each team's progress against our dept goals.

Incentivized

• We pay over $5K/ month and we have high expectations for service. Sometimes I feel that we don't get the value, but only sometimes.
• We have had to build our own application to keep state and broker releases and deployments. We call our app deployer. I feel that CircleCI could do more to understand our needs and possibly build additional features that would enable us to invest less in build and deployment infrastructure and justify paying more for Circle.

Incentivized

• ROI has been good since it's open source
• Settings.xml need to be backed up periodically. It contains all the settings for your pipelines! We accidentally deleted before and we have to restore and re-create several missing pipelines
• More straight forward use of API and allows filtering e.g., pull all pipelines triggered after this date

Incentivized