Trilio Extends Disaster Recovery Reach to Red Hat OpenShift Virtualization
Trilio is making available a technology preview of an instance of its disaster recovery (DR) platform that supports Red Hat OpenShift Virtualization, which enables IT teams to encapsulate virtual machines in a container.
Trilio Site Recovery (TSR) makes use of continuous replication to automate failover and failback operations across sites, a capability that can now be extended to legacy monolithic applications deployed using Kernel-based virtual machines (KVMs) deployed on an instance of a multi-tenant Kubernetes cluster using open source KubeVirt software.
David Safaii, chief evangelist for Trilio, said this latest addition to the Trilio portfolio expands the range of cloud-native services that IT teams can support using a DR platform that is not tied to a specific storage platform. That latter capability is especially critical because in the event of an outage it provides IT teams with a more flexible approach to DR spanning multiple platforms, noted Safaii.
It’s not clear how many IT teams are using the Red Hat OpenShift platform based on Kubernetes to run legacy monolithic workloads on KVMs that have been encapsulated in a container. However, as more cloud-native applications are deployed, many organizations are looking to centralize the management of IT by deploying multiple classes of workloads on Kubernetes clusters. The overall goal is to reduce the total cost of IT by eliminating the need to manage legacy infrastructure.
The challenge is that as IT teams adopt Red Hat OpenShift Virtualization they run into a potential DR issue that Trilio now solves by enabling them to define policies based on the requirements specified in an existing service level agreement (SLA), noted Safaii.
Designed to run on Red Hat OpenShift 4.2 or a higher release of the platform, IT teams can also use TSR to validate their DR recovery plans in fully isolated environments without impacting production workloads, he added.
Red Hat has been making a bigger push for running virtual machines on Kubernetes clusters following the acquisition of VMware by Broadcom. At the core of that effort is KubeVirt, which, once a virtual machine is converted to a KVM, can be used to run a legacy monolithic application at a lower total cost. Each platform an IT organization needs to manage invariably increases the total cost of IT, which in part helps explain why there is a growing interest in platform engineering as a best practice for managing DevOps workflows at scale.
However, many organizations are not going to make that transition unless there is a means to ensure that data and workloads can be recovered in the event of a disaster, said Safaii.
The challenge, as always, is getting the various factions within the IT organization to come to some type of consensus on just how many platforms are actually required. As every IT professional well knows, that decision can be driven as much by cultural preferences as much as it is actual economics, but one way or another, the longer term trend usually tends to favor eventual consolidation.
