As far as data crises go, Nirvanix is probably the worst we’ve seen in a long while. It’s even possible that we could soon see the addition of a new word to the IT lexicon: Take care in choosing a cloud provider so you don’t get Nirvanixed.
The San Diego provider announced earlier this month that it is closing shop on September 30, giving hundreds of users, including some high-profile names like NASA, Fox Sports and National Geographic, until mid-October to pull all their data out or risk losing it. The company apparently lost a critical round of financing as investors increasingly came to the conclusion that it could no longer compete against larger rivals. Late word has it that the company could still pull through – if only because top platform providers like IBM and Intel steer some of their systems customers to Nirvanix for cloud backup, and they would hate to see their reputations tarnished as well. As of yet, however, no lifelines have been announced.
In a sign of how cutthroat the cloud industry has become, rival providers wasted no time reaching out to bewildered Nirvanix customers. While many observers used the episode as a chance to harangue the cloud industry over reliability and availability yet again, the fact is that when large volumes of storage are needed in a moment’s notice, few options exist other than to provision new cloud resources— even if the cause is the demise of an existing service. And with reports that some organizations are looking at preserving 20 petabytes of data or more, there simply isn’t enough time in this case to build new storage infrastructure in-house.
Of course, finding somewhere to put your data is relatively easy. The real challenge is the migration. As InformationWeek’s Charles Babcock points out, there is only so much network bandwidth servicing Nirvanix data centers, which is fine when customers need to slowly add data to backup infrastructure over the years. But with everyone pulling all their volumes out at once, it’s kind of like trying to evacuate New York City in the face of an oncoming tidal wave: The infrastructure just isn’t enough for all that traffic. As Gartner analyst Kyle Hilgendorf puts it, this is not the time for Nirvanix customers to launch a sober assessment of their options but to “react … and react in panic.”
For those fortunate enough not to get caught up in this saga, Nirvanix is about as clear a cautionary tale is you can get. Nicos Vekiarides, CEO of cloud provider (and Nirvanix rival) TwinStrata, notes that the best offense in a situation like this is a strong defense. That means not only ensuring adequate bandwidth for both normal and emergency situations, but implementing multiple backups across various services and then ensuring everyone has compatible standards, formats and protocols to ensure that data can be quickly accessed even if one link in the chain goes down. Trustworthiness in a cloud provider is paramount, but let’s face it, sometimes the unexpected happens, which is why good planning is always the most valuable tool in the IT shed.
The cold, hard fact is that Nirvanix will not be the last cloud provider to leave customers in a lurch. As reliance on the cloud grows, failures will increase—and not all of them will be technological in nature. The enterprise industry could react by retreating back behind the firewall, but that will ultimately cost more in lost productivity than it saves in risk avoidance—internal infrastructure has a habit of going down, too.
If Nirvanix does ultimately pull the plug on itself, the best thing to do is learn from the experience and then apply that knowledge to future cloud deployments.