News Stay informed about the latest enterprise technology news and product updates.

Cell phone carrier picks startup Continuity Software for DR testing

When a cell carrier tried Israeli startup Continuity Software's RecoverGuard, it found problems it didn't suspect it had. Now it uses the monitoring app for constant DR testing.

Can a software vendor that specializes in delivering bad news become an established data center play?

Israel-based startup Continuity Software is trying to accomplish that with its RecoverGuard disaster recovery (DR) monitoring and reporting software. RecoverGuard uses what Continuity calls a "gap detection engine" to validate that applications are being replicated and can be recovered in case of a failure. Continuity offers customers a 48-hour trial to assess its network, and CEO Gil Hecht said RecoverGuard finds problems nearly every time.

More stories on disaster recovery
DataCore adds disaster recovery to virtual iSCSI SAN

Riverbed adds disaster recovery features to WAFS

Data storage startups emerge from stealth
"When they came in, I told them, 'Get out of here,' " said Shlomi Penner, storage system architect and manager of Israeli cell phone company Partner Communications Co. "I said we have no problem in this environment because we are very strict and straightforward. There are no problems here. Then they asked to run their software trial, and they actually found some interesting things going on that weren't acceptable."

Penner ran RecoverGuard on a handful of major databases -- his billing system, ERP and CRM. He said he found a few problems and decided to install the agentless software across his entire data center. Partner Communications' data center is no trivial operation. It includes six EMC Symmetrix DMX systems in three locations with 1.3 PB of storage and 800 ports on Cisco Systems Fibre Channel SAN switches.

RecoverGuard found problems that Penner probably never would have discovered until they caused a failure, he said. Some require immediate attention, while others can wait a few days. But, he noted, RecoverGuard "doesn't fix the problem, it sends pop-up alerts and gives us reports for each week about what alerts are new and what alerts to take action on. We can set an alert for a known issue."

Before installing RecoverGuard, Partner Communications used to manually test its disaster recovery "when we had time," Penner said. "Now we do it on a daily basis."

RecoverGuard has detected a variety of problems over the year in which Penner has used it. "If we have a LUN connected to more than one host and the hosts don't have a cluster between them, then it's a problem, he said. "On these we take action immediately. For disaster recovery problems, such as one LUN in a volume doesn't have replication, then we don't have backup with replication. We can take a week to solve that."

Penner said he likes the fact that RecoverGuard is agentless because he only needs to install an agent on one host. He also likes its new GUI, mainly because he suggested that Continuity improve its original GUI. "Anything we asked them to do, they did it in a week or two," he said.

It helps that Partner Communications is an EMC shop. RecoverGuard only supports EMC and NetApp storage so far. EMC sells it as a services offering in Israel, resulting in a big chunk of Continuity's sales so far. Because he uses Symmetrix, EMC Control Center for SRM and SRDF for replication, Penner falls in RecoverGuard's sweet spot.

But not every company does. "Continuity Software goes deep, but not wide," said analyst Greg Schulz of the StorageIO Group. "They're focused on replication, and they can manage data replication and data protection all the way back up to an Oracle table and all the way down to a particular volume on a particular storage array. But because they go deep, they only support so many platforms."

Hecht said he intends to add more platforms, but the latest release of RecoverGuard last October addressed adding features, such as a configuration wizard, a dashboard and the new GUI.

"It's a nice-to-have now," Schulz said. "If they were ubiquitous across all platforms, it can go from a nice-to-have to a need-to-have."

As Penner said, "It's one step, and it's easier to find the problems than fix them."

Dig Deeper on Disaster recovery planning - management

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.