You are viewing version 2.26 of the documentation, which is no longer maintained. For up-to-date documentation, see the latest version.
What is Spinnaker?
Watch Ethan, an Armory engineer, explain Spinnaker in three minutes.
Spinnaker is an open source, multi-cloud Continuous Delivery and Deployment platform that provides a single pane of glass with visibility across your deployment for deployment status, infrastructure, security and compliance, and metrics. By using pipelines, flexible and customizable series of deployment stages, Spinnaker can fit a variety of deployment needs.
Spinnaker can deploy to and manage clusters across Amazon Web Services (AWS), Kubernetes, and Google Cloud Platform (GCP). More functionality is being added all the time. For example, work is being done to improve integrations with all cloud the providers, including Cloud Foundry and Microsoft Azure.
Spinnaker not only enables businesses to move to the cloud but makes it easier for them to adopt the cloud’s advantages.
Safety and speed
Today’s world revolves around software and services working reliably and continuously – the internet is accessible 24 hours a day, and users expect 100% uptime. The cost of business services experiencing downtime, planned or unplanned, is only growing. Businesses need to be able to deploy software in a safe way with velocity.
Shipping changes more frequently allows developers to gather real user feedback sooner, enabling them to iterate and build based on actual input from customers. Additionally, Spinnaker abstracts away much of the cloud configuration details, giving developers more time to focus on meaningful tasks instead of infrastructure details.
In the past, releases were large monoliths, and ensuring uptime (or deployment safety) meant a long wait between each release, including maybe even extended code freezes. A company that wanted to maintain a stable environment could become averse to pushing out new features, leading to a slowdown in innovation. This tradeoff is getting more and more difficult to justify. To thrive, businesses need a way to deploy software with velocity.
Spinnaker solves these problems by enabling safer and faster deployments with the following benefits:
Immutable infrastructure that builds trust by making sure infrastructure matches an understood and explicit pattern that does not change once it is deployed. If changes are required, a brand new instance gets deployed. Having unique instances for each build enables the use of different deployment strategies, which are another benefit of Spinnaker.
Deployment strategies to fit your needs and infrastructure. The strategies include the following:
- Blue/Green: An easy way to think about this is that it is similar to active-active high availability. You have two instances of your deployment running concurrently, the production build and a new one. Once you feel confident that the newer build is stable, traffic is shifted all at once from the old deployment to the new one. A configurable number of server groups are maintained, which allows for easy rollback in case of issues.
- Rolling Blue/Green: Similar to Blue/Green, but traffic is gradually shifted from the older deployment to the new one.
- Highlander: Similar to Blue/Green, except the old deployment is destroyed once traffic is shifted.
- Canary: This consists of three instances: the current production instance, a baseline instance (a smaller clone of production), and a canary instance with the new deployment. The production instance handles most of the load while the baseline and canary each receive a smaller amount. After a predetermined amount of time, performance of the baseline and canary are compared. Whether the a deployment becomes the new production build depends on canary analysis that can be automated or manual.
Automated canary analysis through Kayenta, a canary analysis tool that is integrated with Spinnaker. Without manual intervention, Kayenta can determine if a canary deployment should be pushed to production.
Multi-cloud deployments to avoid lock-in and allow you to optimize for things like cost, latency, and geographic distribution.
What a typical workflow looks like
A typical workflow with Spinnaker starts with baking a Linux-based machine image. This image along with your launch configurations define an immutable infrastructure that you can use to deploy to your cloud provider with Spinnaker. After the deployment, run your tests, which can be integrated with Spinnaker and automatically triggered. Based on your deployment strategy and any criteria you set, go live with the build.
Armory’s platform includes an enterprise-grade distribution of Spinnaker that forms the core of Armory’s platform. It is preconfigured and runs in your Kubernetes cluster. The platform is an extension of open source Spinnaker and includes all those benefits as well as the following:
- Pipeline as Code (Dinghy) allows you to store Spinnaker pipelines in Github and manage them like you would manage code, including version control, templatization, and modularization. Spinnaker pipelines are flexible and customizable series of deployment stages. Combine all these to rapidly and repeatably scale pipelines in your Spinnaker deployment.
- An Armory extended version of the Spinnaker Operator helps you configure, deploy, and update Spinnaker on Kubernetes clusters. If you cannot use a Kubernetes Operator, you can use Armory-extended Halyard to accomplish the same tasks.
- Policy Engine helps you meet compliance requirements based on custom policies you set.
- Integrations with many of your existing tools, such as Terraform. For a full list, see Integrations.
The services that work together in Spinnaker™
Halyard is a versatile command line interface (CLI) to configure and deploy Spinnaker™.
Learn how Fiat manages permissions in Spinnaker™.
Learn how to control ingress in Spinnaker with load balancers.
Create an application in Spinnaker.
Create your first pipeline, which bakes an Amazon Machine Image (AMI).
This glossary is a list of words and phrases and their definitions as they apply to Spinnaker.
Was this page helpful?
Thank you for letting us know!
Sorry to hear that. Please tell us how we can improve.
Last modified January 22, 2021: (f11cde0)