HCX DR – Not Only for DR

HCX is well known as a migration tool for moving VMs and extending networks between two sites (on-prem DC’s or cloud (VMC, GCVE or AVS), but it also has a less well-advertised feature — Disaster Recovery. HCX DR is sometimes called the “poor man’s DR” because it lacks orchestration (although it can be integrated with VMware SRM for orchestrated recovery and the ability to run scripts after failover). Despite missing orchestration, the replication feature works well, and under the hood, it uses the same vSphere replication technology, giving us solid RPO (up to 10 mins), failback, and multiple restore points.

These can be helpful in several enterprise scenarios:

1. Migrating large production workloads where extended testing is needed. In the medium to large enterprises, any migration is taken seriously, with extensive approval chains. All possible scenarios must be covered, and proof provided that the system will work in the new environment before approval to execute the migration is granted. Usually, the HCX migration toolkit is not suitable in cases where you have to run an isolated copy on a new hosting environment, and the original VM(s) must continue running and serving existing workloads. It is theoretically possible to clone the VMs and migrate the clones, but for larger and especially database VMs, there is often not enough free space to complete the cloning. Instead, we can use HCX DR and, after initial data is replicated, make a test recovery, which will create a clone from replicated data and make it possible to perform all necessary tests .
2. Faster failback after migration with no data loss. During complex migrations, especially for business-critical systems, there is a short, strict migration window within which the migration needs to happen, including all verification testing and — in case of testing failure — failback. For production workloads, final testing is done using live data, and in failback scenarios, that data has to go back, too. Using the HCX migration toolkit means that the VM migration process starts from scratch, this time from cloud to on-premise – and, in the case of large VMs, can take many hours to complete. Meanwhile, using HCX DR for migration after failover reverse replication can be set up with only a few steps, and failback can be performed in minutes.
3. Bubble testing of software upgrades. Not all enterprises have the luxury of having multiple environments for their business-critical systems, where test/ UAT reliably matches production and all OS/software patches and upgrades can be tested with confidence. HCX DR makes it possible to make test recoveries on isolated network segments, and to test out any updates or configuration changes on a clone production system, avoiding unpleasant surprises when doing changes on live production systems.

This blog was contributed by:

Justas Krikštaponis, Chief Technology Officer, TeraSky Baltics