“High Availability is fundamental to enterprise customers looking to run production workloads on OpenStack. Given the massive demand we’ve seen from enterprise customers, we are excited to provide a solution with the industry’s first High Availability capability for both traditional and scale-out workloads,” said Madhura Maskasky, co-founder and vice president of product at Platform9. “This new capability makes it possible for customers to migrate to OpenStack for mission-critical workloads without sacrificing the powerful, enterprise-grade High Availability capabilities they’ve come to expect.”
“For production workloads, High Availability is a critical capability and we have had to run those workloads on VMware vSphere thus far. With this update to Platform9 Managed OpenStack, we can now use OpenStack and KVM for all workloads,” said Rob Horstmann, Manager of Technical Operations at Moz, Inc.
Traditional applications need support from the infrastructure layer to be highly available. Historically this capability has only been available in VMware vSphere and not with OpenStack. Because of these limitations, enterprise customers have found it challenging to run production workloads on OpenStack, and instead have relied on expensive VMware ELAs for such scenarios.
Likewise, cloud-native workloads need support from the infrastructure layer. While these workloads are designed to “scale-out” across multiple-nodes, they are only highly available if the infrastructure is distributing the worker nodes across failure domains (such as hosts, racks, or data centers). Until now, there has not been an out-of-the-box solution for workloads that require both programmatic scale-out (using auto-scaling-groups) and High Availability awareness.
Today, cloud architects can use the concept of availability zones — already in OpenStack — to automatically configure groups of servers that represent a zone of availability, such as a rack or a data center. Platform9’s newly announced High Availability capability automatically configures liveness detection among servers that are in such a zone. When a server in a zone fails, the capability orchestrates the recovery of workloads running on that server onto other servers in that zone. In addition to recovering traditional workloads, scale-out workloads that are being spawned by auto-scaling-groups can be distributed across availability zones, mitigating the risk of simultaneous failure.
This High Availability capability is available in Platform9 Managed OpenStack Enterprise Edition. Platform9 is also open sourcing this capability, making it freely available via a Github repository where the code, documentation and demonstrations reside.
Additional Resources
- Read the blog by Madhura Maskasky: https://platform9.com/blog/
virtual-machine-high- availability-with-platform9- openstack - Join Platform9’s upcoming webinar discussing business value and architecture for VM High Availability in OpenStack: https://cc.readytalk.com/
registration/#/?meeting= smc2ron7t05w&campaign= ivnj2lpm7e1l
About Platform9
Platform9’s mission is to make cloud infrastructure easy for organizations of any scale. The company’s service transforms an organization’s server infrastructure into an AWS-like agile, self-service cloud within minutes, enabling organizations small or large to run their cloud infrastructure efficiently, at scale, while leveraging the latest open source innovations. Powered by OpenStack and Kubernetes, the service is the industry’s first 100% cloud managed platform for KVM, VMware vSphere and Docker. Founded in 2013 by a team of early VMware engineers, Platform9 is backed by Menlo Ventures and Redpoint Ventures and headquartered in Sunnyvale, Calif. For more information, visit: http://www.platform9.com.