Colleges and universities have many options when it comes to a data backup and disaster recovery strategy, from building in-house secondary data centers and using co-location facilities to taking advantage of the latest disaster-recovery-as-a-service offerings. I explored the pros and cons of each and everything in-between – and the cloud won out.
In my tenure as a CIO, I’ve seen disaster recovery plans that were extremely limited and virtually non-existent. I’ve been exposed to previous IT administrations that only backed up the core ERP system, which included the student information system.
That amounted to backing up just 10 percent of the university’s overall data, which was problematic for two reasons: backing up only a small subset of data did not meet data retention requirements, which are typically three to five years. In addition, we had no backup servers and other IT infrastructure in place to bring applications back online. The process of buying new servers and storage hardware, installing them and getting our systems back up and running would have taken weeks.
I always want a disaster recovery plan that can back up all our data and allow us to quickly bring up systems in the event of a natural disaster or other emergencies that would cause our data center to go down.
Here are the five reasons turning to the cloud for data backup and disaster recovery was the right solution for my campus:
1. It’s cost-effective : I explored several disaster recovery options. Historically, doing it in-house is more affordable. Building a secondary data center on or near our campus was not an option because a tornado could wipe out both facilities. I looked into co-location facilities, but their prices were exorbitant. I even talked to colleagues at other institutions in North Carolina and New York about trading rack space – and being each other’s backup data centers, so we can all have redundant systems.
Initially, trading rack space with another university was our most affordable option, but then we talked to Barracuda Networks, who offered us an amazing deal for their data backup and disaster recovery cloud service.
It was an opportunity we couldn’t pass up. First, Barracuda offered us unlimited data cloud storage. (And as a result, it’s allowed us to retain all our data for seven years, which exceeds all legal requirements.) And secondly, by choosing the Barracuda cloud, it included everything we needed, and we didn’t have to purchase any additional data center equipment, which we would have had to do with the other disaster recovery options.
Our three-year subscription fee for the Barracuda cloud service is three times cheaper than if we traded rack space with another university and five times cheaper if we used a co-location facility.
2. Fast Recovery Time : I’m a big believer in server virtualization and virtualization made it very easy to deploy disaster recovery in the cloud. Now, if our data center is incapacitated, we can re-launch all the virtual servers in the Barracuda cloud and bring our applications back up within a few minutes.
During the process, we switch IP addresses. Once we provide administrators and other key personnel the new IP addresses, they can immediately access applications and data over the cloud.
In contrast, if we traded rack space with another university, it would take us a few hours to stand up the servers and bring our applications back up. That’s because not many universities would allow us to do a hot-hot configuration, where our servers and data would be replicated from our data center to their data center in real-time. Doing so would require a lot of bandwidth, and not many universities would be willing to give up that amount of bandwidth 24/7.
3. Multiple points of failure : By going to the cloud, we eliminated the single point of failure. To implement our new disaster recovery solution, we purchased a Barracuda Backup appliance that backs up our virtual servers and data locally to disk. And then, the data is replicated to the cloud.
4. Massive time savings : We don’t have to maintain, secure or regularly update a secondary set of servers and storage in a secondary location. This frees up the IT staff’s time, so we can focus on other important needs such as ensuring all our existing IT infrastructure is secure.
5. It’s also easy to implement and use : During the initial setup, it took a couple of days to back up 120TB of data from our data center to the cloud during off-peak hours. We back up incrementally throughout the day – a process that takes just a few hours. It’s all automated. In fact, anytime we launch a new application on a new virtual server in our data center, the virtual server is automatically replicated on the Barracuda Cloud.
And if our main data center goes down, we could also configure our system to automatically failover to the cloud. But we’ve chosen to do it manually. It’s simple to do. We sign into Barracuda’s secure website and pick the virtual servers we want to re-launched. It just takes a few clicks to stand up for a virtual server in the cloud.
Overall, disaster recovery in the cloud is easy to set up and it’s easy to use. Everything has been smooth. Thanks to the cloud, we now have a worry-free disaster recovery.