Michael Flippo - stock.adobe.com
Engineers slash RPO/RTO as they swap tape for Rubrik backup
Frankham Group ditches archaic tape backup and deploys Rubrik backup appliances and remote disaster recovery facility. Meanwhile, the engineering group assesses a move to the cloud
Kent-based construction design consultancy Frankham Group has ditched tape-based backup and deployed Rubrik appliances in a move that has seen it slash data restores from days to hours and gain the use of full disaster recovery (DR) provision when needed.
The Sidcup-based company – which is considering a move to cloud-based IT (see box) – has 200 employees and four sites including its HQ-based datacentre, from where it runs a Nutanix hyper-converged infrastructure and a suite of Autodesk and Trimble design applications, as well as office products based around Microsoft 365.
About 50% of the company’s teams need to collaborate on design applications, while surveyors tend to use less resource-hungry products.
Previously, backup software was provided by Veritas and was tape-based in terms of media. Tapes had to be taken to a second site in Haywards Heath, around 40 miles away, where they were kept and from where restores took place via previously used but end-of-life vSphere virtual servers.
Andy White, group IT director at Frankham, said: “By the time tapes had been taken to Haywards Heath and restored, it was four or five days behind. We never had data available that was recently saved.”
Also, said White, the integrity of backups was never really confirmed, and in some cases, tapes didn’t get to the second site unless a member of staff happened to be going there, which was a risk because it meant production and backup copies were at the same site. Additionally, the nodes from which data was restored were old and unreliable.
To address this, Frankham Group embarked on the deployment of Rubrik backup software and appliances plus disaster recovery failover to its backup partner’s site, Assured Data Protection in Leeds.
Read more about backup and disaster recovery
- Backup maintenance – make sure backups can deliver: Deploying backup but overlooking the need to make sure it works is a common error. We look at the why and how of backup maintenance to help ensure you can recover from your backups.
- SME disaster recovery – five key points to consider: We look at key DR considerations for SMEs, including RPOs, RTOs, infrastructure needed for effective continuity and the right combination of cloud and in-house provision.
Rubrik appliances are deployed as 2U 20TB nodes at Sidcup with nightly backups. Thereafter, a synch to Assured’s infrastructure takes place. Regular testing of the DR capability is also carried out.
Key benefits for White are peace of mind and the rapidity and reliability of restores.
“We now have no worries about whether backups are working,” said White. “Often previously backups were not verified. Now our partner monitors backups and the capacity needed to keep them and we can get daily reports on each server. We also have the flexibility to back up from an entire server down to virtual machines and drives.”
White said recovery point objectives (RPOs) and recovery time objectives (RTOs) had been slashed from up to a week down to a few hours, and that the DR suite could be spun up in a few hours if needed.
Frankham Group eyes the cloud but with reservations
Frankham Group is considering a strategic move to the cloud, but has serious reservations about cost and user experience. Its concerns typify those of a smaller organisation that wants the benefits in terms of simplicity the cloud could provide, but worries about cost and the effect on user experience.
It currently runs four Nutanix hyper-converged infrastructure nodes that provide compute and storage from its Sidcup headquarters. These provide Autodesk and Trimble applications to engineers, who often work remotely. Key to providing these applications is to ensure user experience because compute and graphics requirements are high. That’s why on-premise works for Frankham right now.
But Frankham’s group IT director, Andy White, said some of its competitors had moved to the cloud, and it was something he was looking at.
“We will stay as we are currently, then when that cycle is over, we will see if we can transfer to the cloud, said White.
“But the end user experience has to be good. We have one year left on the Nutanix contract, but there’s no real rush. We get what we need in terms of end user experience, so that’s the main thing, but we will want to check out cloud options soon.”
Cost is also a consideration, said White.
“Last time we priced things up with the Nutanix reseller, the best bet in terms of user experience and cost was to have data at our datacentre,” he said.
“But we want to take away the need to manage things at our offices and put services in the cloud. If it was just the Microsoft 365 applications, it’d be an easier decision, but with Autodesk and Trimble applications, compute and graphics requirements are more rigorous.”
