As part of XIOS 6.1 which contains the new and revolutionary native replication (I blogged about it here ..), we have also released the support for VMware site recovery manager (also known as SRM).
You wouldn’t drive a racecar without a helmet… and you shouldn’t run your data un-protected.. Especially as you consolidate more and more mission critical workloads! But not just any BC/DR and data protection solutions will do – XtremIO requires flash optimized solutions that can protect workloads at the speed of flash. XtremIO delivers best-of-breed scalable native integrations, for homogenous or heterogeneous DR storage, across local and remote, and asynchronous and synchronous solutions.
Built intro XtremIO are native snapshots called XVCs for local data protection and snapshotting.
Inside the data center leverage Data Domain with ProtectPoint Technology
And across the Data center for asynchronous local, remote, and virtualized protection native offloaded replication with Dell EMC Recoverpoint.
Synchronous local and metro continuous availability replication is delivered via Dell EMC VPLEX.
Up until now, the only way to replicate data between sites was to use the physical or the virtual edition of Recoverpoint, this is now changing with XIOS 6.1 that provides a truly unique, meta-data aware replication technology
VMware Site Recovery Manager
vCenter Site Recovery Manager is a disaster recovery and business continuity solution with the following characteristics:
- Operates as an extension of vCenter Server
- Automates the recovery or migration of virtual machines between a protected site and a recovery site
Uses array-based replication solutions from other vendors, in our case, its XtremIO
Planned migration is a vCenter Site Recovery Manager workflow that enables application-consistent migration of virtual machine workloads with zero data loss to the recovery site
Automated failback provides an automated workflow to migrate virtual machines back to the protected site after the recovery site is restored.
About Storage Replication Adapters and Array Managers
An SRA is a program provided by a storage array vendor that enables vCenter Site Recovery Manager to work with the vendor’s array.
Functions performed by SRAs:
- Array discovery
- Replicated logical unit number (LUN)
- Initiation of recovery plan tests and
An SRA is used by an array manager, a vCenter Site Recovery Manager component that helps to identify available arrays and replicated LUNs.
You test a VMware vCenter™ Site Recovery Manager™ recovery plan by simulating a failover while isolating the test environment from the production environment.
You protect the new protected site after a recovery using reprotect.
You restore the initial vCenter Site Recovery Manager configuration for protected and recovery sites using failback.
Failover and Planned Migration with vCenter Site Recovery Manager Recovery Plans
Unplanned Failover – Recover from unexpected site failure:
- Full or partial site failure
The most critical but least frequent use case:
- Unexpected site failures do not happen often.
- When site failures happen, fast recovery is critical to the business.
Preventive Failover – Anticipate potential data center outages:
- Examples: hurricane, floods, forced evacuation, and so on
Initiate preventive failover for smooth migration:
- Graceful shutdown of virtual machines at protected site
- Using vCenter Site Recovery Manager planned migration capabilities to ensure no data loss
Most frequent vCenter Site Recovery Manager use case:
- Planned data center maintenance
Global load balancing
Ensure smooth site migrations:
- Test to minimize risk.
- Execute partial failovers.
- Use vCenter Site Recovery Manager planned migration to minimize data loss.
Use automated failback, which enables bidirectional migrations.
Planned Migration and Application Consistency
Planned migration ensures application consistency and no data loss during migration by providing the following:
- Graceful shutdown of production virtual machines in an application-consistent state
- Data sync to complete replication of virtual machines
- Recovery of fully replicated virtual machines
Recovery Testing and SRM
The ability to non-disruptively test a recovery plan is one of the most powerful features of VMware Site Recovery Manager (SRM). Frequent testing reduces risk by reducing drift between plans and desired behavior. It also provides the confidence that if a disaster strikes you are ready, knowing how SRM and your applications will respond.
An SRM recovery plan test is designed to be as close as possible to running an actual recovery plan while not disrupting the protected VMs and not impacting replication.
From a storage perspective, SRM will utilize the XtremIO snapshots at the designated testing site so replication will still keep going on and from a network perspective, you can either designate your own dedicated vSwitches that needs to be isolated from your production network or you can use the built in SRM “auto-network”, while it’s easier to go with the latter, networking will only work between the VMs running on that host during the test since, its not connected to any physical nic so really, the latter option is more of a safety belt mechanism rather than a true test scenario so you really want to designate a real, isolated network vSwitches for the test.
Ok, now that we cover the main highlights of VMware SRM, let’s see how it all connects together, instead of just flogging you with more details, I created a series of videos showing you
- XtremIO native replication overview
- Performing SRM test failover and cleanup
- Performing SRM failover, reprotect, failover (failback) and reprotect
but wait, there’s more..
VMware SRM is awesome but it has one big limitation, no matter how many points in-time your array support natively, if you use SRM, you can only failover or test-failover to the last point in-time…well, shortly after the release of XIOS 6.1, native replication and the SRA adapter, we will also release a new version of our Virtual Storage Integrator (VSI) plugin for vCenter which will allow you to select the PiT you want SRM to use during the failover/ planned migration/ Test failover, below you can see how it’s going to look like and of course, i will put up a blog post once it’s out..