News Stay informed about the latest enterprise technology news and product updates.
This article is part of our Essential Guide: The state of the enterprise cloud and prepping for AWS re:Invent 2013

Nirvanix cloud collapse seen as a canary in the coal mine for IaaS

Following news of the Nirvanix collapse, IT pros say a multicloud strategy is needed to protect against cloud vendor failures, despite added costs.

A major cloud storage service provider has abruptly closed up shop, according to reports, leaving cloud users to...

hedge against a similar disaster in other types of cloud services.

San Diego.-based cloud storage provider Nirvanix told customers and partners this week that data should not be replicated to its service any longer and that existing customers must move data off of the cloud storage service by next month, according to a TechTarget report. Nirvanix has declined to comment.

Everyone needs a backup plan.

Richard Calmas,
CEO, Neighborhood Pay Services

This cloud service shutdown is a fairly spectacular one -- Nirvanix has collected some $70 million in funding since its launch in 2007 and is an IBM SmartCloud Enterprise storage services provider. Some reports had hundreds of customers and dozens of petabytes of data under management. The company's product viability was also labeled "excellent" by Gartner Inc. in a December 2012 report comparing public cloud storage vendors.

This scenario gives IT pros that rely on services within the evolving cloud market reason for concern. In an Infrastructure as a Service market, where Gartner estimates Amazon Web Services (AWS) has more than five times the compute capacity in use than the next 14 providers combined, it's easy to surmise that some of the smaller IaaS players won't make it long-term.

Prepared for dissipating clouds?

Some enterprises use small IaaS providers, citing factors such as ease of use or greater flexibility in deploying resources than AWS offers. What will they do if one of these providers goes under?

"We believe in redundancy," said Richard Calmas, CEO of Neighborhood Pay Services LLC, a financial service provider for the housing industry based in Newton, Mass., which uses IaaS services from AWS and smaller provider ProfitBricks GmbH.

The company had used AWS alone, but ran up against problems with its user interface and customer support, prompting it to switch its primary IaaS location to ProfitBricks, which is headquartered in Germany but has a U.S. location in Cambridge, Mass. But AWS is still used for backup to hedge against just this situation, Calmas said.

It does add cost to sync the same data between two clouds at once, but "it's the cost of doing business," Calmas said. "Everyone needs a backup plan."

The belt-and-suspenders approach of a mix of cloud and local resources has worked for Brickfish, a Chicago-based social media marketing software provider. CEO Michael Mullarkey said his company uses multiple cloud providers but conducts its business primarily on NaviSite's cloud. The company also performs local backups, which are sent to Iron Mountain for disaster recovery purposes.

"One of the benefits of cloud is that you can take a snapshot of what you have and virtualize it somewhere else pretty quickly," Mullarkey said.

Still, in the worst-case scenario in which data is lost, Mullarkey said he's keenly aware that clients aren't going to blame cloud providers such as Rackspace, Amazon or NaviSite.

"They're going to look at us," he said.

In addition to IBM, Nirvanix had a number of reseller deals with companies, so organizations might not even have known its data was stored with Nirvanix until this week, pointed out Carl Brooks, analyst with 451 Research, an IT research firm based in Boston.

"White-label partners of Nirvanix might have been banking on Nirvanix and not telling anybody," Brooks said. "If customers aren't able to know the full chain of custody of their data, they could be in serious trouble."

Brooks advises careful evaluation of data that is put into the cloud, as well as thorough due diligence in finding out which partners a company works with to provide a service.

In the meantime, Brooks said, cloud storage services have matured at a faster rate than IaaS, where there are still debates about things such as OpenStack, and the industry is still developing. Thus, cloud storage failures -- which have included an EMC Atmos online storage service, as well as sites such as Megaupload -- could be a harbinger of things to come as the IaaS market matures.

"Nobody can treat cloud as a magic cure-all; everything fails," Brooks said.

Beth Pariseau is senior news writer for SearchCloudComputing. Write to her at or follow @PariseauTT on Twitter.

Join the conversation


Send me notifications when other members comment.

By submitting you agree to receive email from TechTarget and its partners. If you reside outside of the United States, you consent to having your personal data transferred to and processed in the United States. Privacy

Please create a username to comment.

Do you have a cloud contingency plan?
We have a single product to provide end-to-end backup, disaster recovery, virtual hypervisor backup and DR as well as cloud support. The data is always accessible from anywhere we have internet access through a simple to use GUI.
We back up all our data in-house in duplicate. We do not use cloud services.
Working on it currently.
We’ve our own local backup.
My own private servers and cloud, with my own backup. External clouds are not confiable at all. And never will, I think. Because we all are too small to the great players (Amazon, Microsoft and so).
We advise and work with our clients to develop a cloud (IaaS, PaaS, SaaS) exit strategy, not just in the event of a failure but also if a superior solution comes to market
The real issue is creating such a plan and keeping it current as the company's cloud infrastructure changes and grows. There is also a lack of standards and guidance as to how to balance the elements within the plan (we use a modified CBA and GQM approach) to determine to how the elements of the plan should be prioritized.
Supplement with private cloud and backup
You can use Ceph open source storage software to create your own low-cost storage in-house on any hardware that can run Linux.
Many solutions now offer the idea of private clouds. I think best solution is a mix as always has been with emerging technologies.
Nirvanix is not a canary in the coal mine. I support the reasons Simon Robinson from 451 research outlined for the failure. Nirvanix had a flawed business model that required too much capital investment and generated too little revenue. Nirvanix is shuttering its service because it cannot raise any more funding after burning through $70M. Nirvanix may also have faced technical difficulties in scaling out their storage service despite Gartner's assurances that their product viability was excellent. Simon Robinson suggests that Nirvanix didn't have much in the way of intellectual property because there was no "fire sale" of the company's IP by Nirvanix. Finally, Nirvanix did not evolve beyond their storage service. Their customers could not grow their use of cloud infrastructure with Nirvanix. A cautionary tale but avoidable if you don't get obsessed with running with the big dogs.
Great tech but flawed biz model. This was a flawed business where they were burning through cash faster than generating revenues. All Clouds are next as a bubble. Even for post IPO companies like Salesforce,( who have been in the biz for ~10 yrs), there's not a dime of earnings. Today the market is valuing these companies on the basis of growth, but day is not far when solid earnings will matter- every biz needs to be run off it's own operations independently and not rely on VC/IPO money always- those are crutches that will give way some day- so bottom line is that cloud companies need to be profitable quickly in line with traditional software co. metrics on profitability else Nirva(nix) will happen 10 times over.That's reality -period!
We need to agree that there is a chance that the Cloud service may fail as a business. Need to start thinking about contingency plan
We are still in design/eval mode for cloud based Infrastructure, but the strategy is to move forward in that direction, one of our principle is rent, buy, built... This would be definitely included in cloud env consideration requirement.
Local and offsite solutions for DR is paramount. Virtual servers allow for P2V and V2V recoveries. Still surprised to hear of weekly tape runs offsite, when CDP keeps the RPO so close to real-time