AWS CodePipeline vs. Spinnaker

CodePipeline is well suited for an already existing AWS-native deployment. It is very easy to connect to existing repos like GitHub enterprise or cloud repos like CodeCommit. Being able to define the process by code (YAML) is a huge benefit for developers who favor that type of deployment setup. The UI is easy to use yet very powerful and customizable. Being able to leverage CloudTrail or Lambda is quite powerful, especially in larger more complex projects. It becomes less valuable with smaller projects or locally hosted deployments that don't get the benefits of a managed service in the AWS ecosystem. However, there are agents that can be run on private servers to allow integration. But naturally, smaller one-off projects benefit less from the automation value derived by CodePipeline.

Incentivized

Spinnaker is a strong solution for Continuous Deployment being able to manage enormous environments in an easy way. Even Spinnaker being able to manage environments based on cloud instances (ec2 for example) I believe it is more suitable for containerized environments. Mainly in Kubernetes where it excels as a reliable and safe tool

Incentivized

• It is reliable and works without errors
• It integrates well with our repository and all other AWS functions as well as our end database

Incentivized

• The immutable way of deployment is one of its greatest advantages.
• We have a strict policy to restock our instances with new images very frequently and this can be done very seamlessly via Spinnaker
• Rollback/ resize of clusters is one of the coolest features of Spinnaker.

Incentivized

• Ease of use - things like CircleCI or other tools are a bit easier to learn.
• Ability to build from more sources.

Incentivized

• Integration with external authentication mechanism is not that simple.
• Its configuration is made by a external tool (halyard). Would be better if we could configure it by using its own frontend.
• Its user interface is some times a bit confusing.

Incentivized

Overall, I give AWS Codepipeline a 9 because it gets the job done and I can't complain much about the web interface as much of the action is taking place behind the scenes on the terminal locally or via Amazon's infrastructure anyway. It would be nicer to have a better flowing and visualizable web interface, however.

Incentivized

Our pipeline takes about 30 minutes to run through. Although this time depends on the applications you are using on either end, I feel that it is a reasonable time to make upgrades and updates to our system as it is not an every day push.

Incentivized

We didn't need a lot of support with AWS CodePipeline as it was pretty straightforward to configure and use, but where we ran into problems, the AWS community was able to help. AWS support agents were also helpful in resolving some of the minor issues we encountered, which we could not find a solution elsewhere.

Incentivized

I felt that, out of the alternatives, AWS CodePipeline was the simplest to setup and most reliable. Since my client's infrastructure was already hosted in AWS, I felt it was a no-brainer. If a client needed a similar solution with on-prem or non-AWS infrastructure, I would probably evaluate a different solution. AWS CodePipeline is pretty tightly coupled with the rest of the AWS ecosystem.

Incentivized

• Pipeline Expressiveness • Self-Service/Override • Visibility of Client Teams • Operability of Client Teams - • High-Quality Integrations (AWS, IHP, Google) • Extensibility – (Ability to add code) • The maturity of Deployment Process • Speed/Ease of Onboarding

Incentivized

• AWS CodePipeline reduced CI/CD pipeline development time by 10% for AWS native application stacks.
• AWS CodePipeline reduced response time to build failures by 3% through SNS integrations.

Incentivized

• It's open source!
• RPM based deployments can be scripted at Maven file itself , so very easy!
• RPM based deployments bundles all code and configs together, so awesome!

Incentivized