Cloud Foundry has built-in health management and self-healing capabilities. But when delivering a Cloud Foundry-based service, it is not sufficient to monitor and heal the Cloud Foundry components, since the end-to-end functionality of the service also depends on the state of the entire infrastructure surrounding and supporting Cloud Foundry. This includes the hardware, virtualization and networking platforms on which it runs, the CF Services available, the user interface, and even the billing systems.
To this effect, at Swisscom we have built an overarching Health Management platform called Orchard, which not only monitors and notifies about failures in our Application Cloud platform, but has the ability to trigger automated responses to heal those failures automatically. To challenge Orchard’s abilities, we have also developed Chaos Heidi, a self-disruption platform that allows us to cause failure scenarios of different types.
In this talk we will describe Orchard and Chaos Heidi, including their motivation and architecture, their current implementations, and our challenges and ideas for the future.
Bill is VP of Engineering at Stark & Wayne where he focuses on helping his teams deliver world-class solutions to cloud-native problems.Prior to joining Stark & Wayne, Bill specialized in helping to start software companies along with helping established companies greenfield brand... Read More →
Diego Zamboni is a computer scientist, consultant, author, programmer and sysadmin who works as a Cloud Architect at Swisscom, where he is building monitoring and self-healing capabilities into Swisscom’s Cloud Infrastructure. He has more than 20 years of experience in system administration... Read More →
