When one of the founders of your company is known as the father of American insurance, it’s probably a good idea to follow his advice—even if the advice is more than 260 years old.
We’re talking about Benjamin Franklin and one of his famous thoughts is: “An ounce of prevention is worth a pound of cure. “
The Philadelphia Contributionship, which Franklin helped found in 1752, is listening to that advice by initiating a disaster recovery and business continuity plan to keep the property insurer up and running in the event of a tragedy.
“We were hit pretty hard by Hurricane Irene (in 2011) like most personal lines companies,” says Mark Montagna, IT infrastructure manager for the Contributionship. “But this particular initiative—from a standpoint of replication and high availability disaster recovery—has been going on for a few years. It wasn’t actually brought on by Irene; Irene just underlined the need for it.”
The Contributionship had just one data center and Montagna explains, “We needed to get the data into another data center so we could bear the brunt of a disaster occurring at our headquarters.”
Like many smaller carriers, the Contributionship is located in the same geographic area as its policyholders as well as its employees.
“An event could affect the business and our policyholders,” says Montagna. “We need to take care of our customers in their time of need to maintain continuity in the event of a disaster.”
When Montagna joined the Contributionship two years ago, he found a company in transition with new leadership. He also discovered the technology in place was not suitable for high availability or business continuity. So, one of the first steps was to begin the process of spreading the risk to an additional data center.
“We pushed some of our production systems, which were located at the headquarters, to a remote data center, but there wasn’t a replication scenario,” he said.
Montagna felt it was imperative to utilize the environment of a remote data center, but from a disaster perspective the company was simply backing up data in the event of a disaster. From there the IT team would go to the recovery center and bring it back online.
“It would require a minimum one to three days to accomplish that,” he said. “Replication became the obvious candidate so we could have our systems ‘warm’ versus coming from a ‘cold’ posture if something were to occur.”
The ability to replicate what is being processed on the servers was unheard of for companies the size of the Contributionship15 years ago, according to Montagna.
“Virtualization opened the doors to replication for us,” he says. “To be able to replicate a virtualized server from one location to another utilizing older technology from our SANS allowed us to replicate entire volumes stored in there. That lowered our recovery time from 72 hours to whatever time it took to build the system around the warm data.”
Montagna claims he spent three months with little success finding a software-based solution that was not snapshot based. Montagna was on the Internet trying to find other alternatives that could match up with the company, the network, and the resources of the Contributionship when he came across Zerto and its virtual replication product.
“It was kind of a love story,” he says. “I didn’t believe it when I saw it.”
Zerto is a young new company and Montagna explains the vendor had to prove itself all the way through the process.
“They were telling me they could do something and I could not find anyone else who said they could do this,” he says. “There are solutions providers who promise the world and you bite. You either take your losses, or you lick your wounds and move on.”
Fortunately, this wasn’t the case for the Contributionship, which committed one tenth of its network to Zerto for a disaster recovery exercise, something the insurer does twice a year.
“We picked some of the more critical back-end administrative systems that our company requires to do business on a daily basis,” says Montagna. “We performed the tests and went through what normally takes 24 hours to get those servers online and did it in 10 minutes. The reason it took 10 minutes was because we did them one at a time. We could have had them all online in one minute.”
Zerto allows the Contributionship to scale in a test mode using virtualization, bring it online, test it, and conduct a proof of concept for the users, according to Montagna. All the while that test is going on, replication continues.
“I still to this day don’t know another company doing this,” he says.
Work remains, but the Contributionship is moving full throttle to add the remaining environment. Zerto, meanwhile is adding features and expanding on what they currently offer.
“We can also use this as a development source,” says Montagna. “Those replicas we copy across the network can stand up as a clone for developers to step up and test applications to see if it works and then destroy the clone.”
When a company has been in business for 260 years and lists Benjamin Franklin as one of its founders, there’s a certain amount of pressure to keep the company successful.
Montagna puts it succinctly: “Don’t screw it up on our watch.”