Information Age: News, analysis & insight for IT & business leaders

After a routine power-down caused a week of disruption, the MS Society took control of its own backup facilities

"It was a real brush-up-your-CV time,” recalls Andrew Young, technical services manager for charitable organisation the Multiple Sclerosis Society. He is describing the time when a routine power-down of the organisation’s storage area network (SAN) caused a week-long disruption to emails and office systems.

“We were testing the building’s power supply,” he says, “so we thought it would be safer to power the SAN down.” An engineer from the company that provided third-party support for the SAN was brought in to do so, but when it was switched back on, the stored data was no longer available.

The MS Society then turned to its backup systems, which were located offsite and also managed by a third party. That was when Young learnt how unstable its storage infrastructure really was.

“The real issue was trying to restore our backups,” he says. “We were using a 1Mb leased line, and we had about a terabyte of live data to recover, and it was clearly going to take weeks.

“We realised that we needed the backup drives to be physically transported [to our premises]. But even when we got them there, with an engineer using a 1GB Ethernet connection, it still took forever because all the data was encrypted.”

In short, the storage infrastructure was ill-suited to disaster recovery.“It’s difficult to quantify the disruption caused by not having access to emails for a week,” says Young, “but it was extremely disruptive.”

Young therefore sought to replace both the backup system and the SAN itself. He considered other third-party services, but also looked into systems that the charity could operate itself.

This led him to consider storage devices based on the iSCSI architecture, including Dell’s EqualLogic range. The company allowed the MS Society to try out one backup array, and Young was struck by how simple it was to operate. “When we looked into how much training would be required [to deploy new systems in-house], the EqualLogic systems seemed extremely easy to use.”

He also says that the support and training made available by the provider allowed the charity to iron out most issues encountered when it later decided to virtualise the majority of its server environment. “Now the amount of time we have to spend [managing the backup array] is negligible.”

That experience led Young to deploy EqualLogic devices to support its SAN. “We had envisioned using the EqualLogic system purely for backup,” he says, “but gradually it has become the centre point of our storage infrastructure.”

Not only is the MS Society now confident that it can withstand a major systems interruption, it has saved around £40,000 a year in the process.