Infinidat Supports vVols Replication

Infinidat has had a long partnership with VMware for almost a decade and with that, a long list of powerful VMware integrations and capabilities, making life easier for virtual infrastructure admins building software-defined datacenters, leveraging the industry acclaimed InfiniBox enterprise storage platforms. In 2021 we released InfiniBox 6.0, which added support for VMware Virtual Volumes (vVols) with two key attributes in mind: (1) ease-of-use, and (2) scale. InfiniBox proved unparalleled scale with 200,000 vVols per storage array (>3x the competition at the time). Infinidat Host PowerTools for VMware (HPT-VM), our vCenter plugin, provides single-click setup of vVols datastores.

Our soon-to-be released InfiniBox 7.1 update adds support for vVols replication, adding disaster recovery for virtual machines running on vVols datastores, including orchestration for VM failover using Site Recovery Manager (SRM):

• InfiniBox 7.1 supports async replication for virtual machines that use vVols datastores.
• The vVols replication relies on the same efficient async replication technology that InfiniBox already provides for volumes and filesystems.
• InfiniBox storage administrators easily set up vVols replication very similarly to other types of InfiniBox replication, and these multiple types of replications can coexist and use the same replication network.

Let’s discuss InfiniBox vVols replication and demonstrate the extreme ease-of-use that allows VM administrators to control the InfiniBox replication.

vVols Replication Responsibilities

One key aspect of InfiniBox vVols replication is a clear definition of responsibilities: What do storage administrators need to do? How much are they expected to be familiar with the applications running in the vSphere environments? Are VM administrators expected to be experts in InfiniBox replication technology and the InfiniBox setups?

A simple way to understand these responsibilities is to think of the InfiniBox storage administrators as providing replication "services" that VM administrators consume. The VM administrators assign individual VMs to be replicated, all using Storage Policy Based Management (SPBM).

Storage administrators manage the replication service.

VM administrators manage the replication contents.

vVols Replication Coexist with other types of Replication

vVols replication uses the same network and inter-system connectivity as other types of replication, such as volume or file system replication.

Storage administrators can use their existing replication Network Spaces or create new ones with the same best practices with which they are familiar. In addition, replication links between InfiniBox storage platforms can be used for all replication types.

vVols Replication Groups

How can VM administrators indicate which individual VMs should be replicated? By assigning them to vVols Replication Groups (RGs).

A replication group is a set of vVols that replicate together to a target InfiniBox storage system. It is the smallest set of vVols that SRM can failover, and would typically contain a set of VMs that need to run together.

InfiniBox storage administrators create replication groups and the replicas that define the target system and the replication frequency. They are also expected to monitor the replication service, and to make sure that the replicas meet the recovery point objective (RPO). The storage administrators will find these activities extremely simple as they are identical to setting up and monitoring other types of InfiniBox replication.

That’s actually all the storage administrator needs to do, create a replication group and its replica. Very simple and very intuitive...

Here's a sample screenshot from the InfiniBox management console:

vVols Replication and SPBM (Storage Policy Based Management)

Before VM administrators can assign VMs to replication groups, they need to create a storage policy with a rule (i.e. a requirement) for replication.

Here's a sample screenshot from the vSphere Client:

Now, when VM administrators create a VM, they can choose this replication-required storage policy when selecting the datastore for the VM:

That’s all the VM administrator needs to do: create a storage policy (which is typically a one-time operation) and then assign VMs to replication groups.

Managing Failover and DR Recovery using SRM

As expected, the entire failover orchestration is performed and managed through SRM (Site Recovery Manager).

The administrator's experience using vVols in SRM is almost entirely identical to the experience of using volumes. Just select Virtual Volumes as the protection group type:

And then select which replication groups will belong to the protection group (one or more replication groups):

From here on, it's all the same: define recovery plans, test them, run them, and re-protect failed-over protection groups.

Site Recovery Manager can manage vVols replication without a vendor-specific Site Recovery Adapter (SRA).

Infinidat’s vVols replication integration builds on our trusted and long-running relationship with VMware. We continue to apply our unmatched ease-of-use of requirements in all aspects of our solutions. We want our integrations to be extremely easy and extremely intuitive and not just functional.

For more information, go to the accompanying blog posts listed below:

• New InfiniBox Capabilities for Modern VMware Infrastructures by Erik Kaulberg
• Infinidat Extends NVMe/TCP to VMware Environments by Tsiyon Sadiky

In addition, the following are blog posts by two of Infinidat’s Tech Alliance partners:

• Confidently Manage Complex Data and Storage Environments with Veritas and Infinidat
• Infinidat Adds NVMe-TCP Certification for VMware