Jul 30, 2020 08:00:00

Lyft openly releases 'Clutch' that can collectively manage infrastructure management such as AWS and Kubernetes with open source

Lyft, a mobility network company, has released ' Clutch, ' which can collectively manage infrastructure technologies for web services such as AWS,

Kubernetes , and Envoy . You can now manage infrastructure technology, which used to be a separate management system, from a single service.

Clutch · An extensible platform for infrastructure management.
https://clutch.sh/

Announcing Clutch, the Open-source Platform for Infrastructure Tooling | by Daniel Hochman | Jul, 2020 | Lyft Engineering
https://eng.lyft.com/announcing-clutch-the-open-source-platform-for-infrastructure-tooling-143d00de9713

Clutch is software focused on managing the infrastructure technologies that underpin web services. Although infrastructure technology has made it easier to introduce new infrastructure, the settings, tools, and logs used by each infrastructure are unique, so increasing the scale of infrastructure management itself. It was difficult to do. Clutch can solve such problems, and by integrating each infrastructure management into one service, infrastructure management can be done intuitively and safely.

The architecture of Clutch is as follows. The back end system by

Go operates cloud computing services and monitoring systems, and the front end developed with React is linked with the back end with Protocol Buffers .

An example of actually using Clutch is also introduced. For example, changing the size of an

AWS EC2 Auto Scaling group traditionally required using the AWS management screen or the command line for AWS.

It looks like this when you change the setting of AWS EC2 Auto Scaling via Clutch. Select Resize Auto Scaling Group on Clutch's front end screen...

Enter the group name and click 'CONTINUE'.

You can change the size of the group and click 'RESIZE'. You can change AWS settings without using the AWS management screen.

In this way, if you use Clutch, you can perform infrastructure management collectively on Clutch instead of a management screen for each service.

It is also possible to notify Slack etc. of who made the setting change and how.

By facilitating infrastructure management, it is possible to provide a platform that facilitates quick resolution of failures, prevention of operation mistakes during maintenance, and development. In the future, a monitoring dashboard and escalation function will be implemented.

The source code of Clutch is published on GitHub, and anyone can use it.

GitHub-lyft/clutch: Extensible platform for infrastructure management
https://github.com/lyft/clutch