How can DevOps teams take continuous delivery to the next step? Part of the answer lies in self healing, which could optimize performance management and monitoring and underpin the software delivery pipelines of the future.
Self healing means the ability of systems or environments to detect and resolve problems automatically. It eliminates the need for manual human intervention, upon which most systems today still rely to fix issues.
Self healing is better than depending on humans for several reasons:
- Automated solutions are less likely to miss issues by mistake, or to make errors when resolving them.
- Self-healing technology is better positioned to fix problems faster than humans because it can react instantly based on rules or machine learning. It doesn’t have to stop and think about the issue before taking action.
- A self-healing environment can scale without limit. This is a key advantage over manually managed environments, whose scalability is limited in some respects by the availability of human admins.
Self Healing and Continuous Delivery
DevOps teams already take advantage of many other kinds of automated tools to help achieve continuous delivery.
They use continuous integration servers to automate tests. They adopt release automation tools to streamline deployment. They implement orchestrators and scripted infrastructure to provision environments automatically.
Yet, when it comes to performance management and monitoring, the typical organization is a far way off from achieving a fully automated process. Even the best application performance management (APM) tools of today are primarily capable only of detecting problems and displaying information about them. They depend on humans to interpret and react to that information.
That dependency is a major obstacle for achieving fully continuous delivery. You can’t release code quickly if your process for finding and solving problems related to the code is mostly manual.
By making better use of self-healing technology, DevOps teams can speed continuous delivery in the following ways:
- Developers and admins can spend less time on monitoring and more time completing other tasks that can’t be automated.
- Ideally, teams will be freer to switch between development environments and frameworks without having to establish a new monitoring workflow. Self-healing tools will adapt to whichever environment they have to work with.
- Because problems can be found and fixed faster, infrastructure utilization can increase. The need for a buffer between average usage and overload can become smaller because automated healing solutions will be able to fix problems before they lead to failure.
- Teams can establish better continuous feedback loops because, by removing the human element from performance and monitoring, incidents can be reported instantly, consistently and in a way that makes the data easily accessible to everyone on the team.
To be clear, the technology for self healing does not yet fully exist. Developers over the past several years have made great strides in automating configuration, service discovery and environment provisioning, But perhaps it’s time to make them a more prominent focus. They’ll enable the DevOps innovations of the future.
— Chris Tozzi
