A total data restore would have taken marketing services company M&C Saatchi Australia up to 24 hours to complete using its old backup system.
Infrastructure manager Ed Barrett said it conducted a business continuity review which found the company would be close to exceeding recovery time objectives in a disaster unless backup procedure changes were made.
"We were using array based snapshots and clones as a primary backup mechanism, but they were clunky to configure and provided limited retention,” he said.
“If we ever had to fall back to tape to recover entire volumes of data, we would have been approaching 24 hours before the job was completed”
Because the company’s designs and campaigns for companies such as Optus take up several terabytes of data, Barrett began looking for a replacement in early 2013.
“We got to the point where we needed to evaluate what we were doing in the next few years. We needed to be confident that we had that capacity in place,” he said.
The company rolled out two Nimble CS260G storage arrays in early 2013. These now contain 37 TB of production data which include Microsoft Exchange emails, SQL databases and production files.
The storage array operates within a VMware virtual environment with 30 virtual machines.
Recovery time objectives are now between four and eight hours, depending on how critical the system is.
- Pental upgrades storage for improved disaster recovery
- SAI Global finds value in ditching tape backups
- Nimble Storage eyes A/NZ market
Barrett added that there are plans to relocate the second storage array to an off-site data centre by the end of 2014.
“This will give us the opportunity to fail-over workloads to another site. Replication is so easy to configure, it’s just a matter of clicking on a couple of buttons.
“If we need more capacity [in the future], it’s easy to achieve. I’m extremely confident that if we suddenly had a huge increase in data, it wouldn’t be a problem,” he said.