Insuring against cloud failures

broken-cloudWe all store (and often share) our information on a cloud. Cloud storage is typically free for limited amounts of memory (which may mean 25 GB, which is by no means a small amount...). When we assign our data to a cloud, we implicitly trust it and assume that it can provide a failsafe service, where we can retrieve all our data any time we want to. Unfortunately, that's not the case: clouds suffer from failures as our personal storage means do. And in some cases network failures add to problem, since we cannot access the cloud in the first place. The availability of cloud storage servers has been shown to be quite lower than 100% (see, e.g., the following papers:  "The availability of cloud-based services: is it living up to its promise?"). If we need to have a storage repository as reliable as possible, we can resort to subscribing to several clouds and store our data on all of them: when a cloud fails, it is likely that some of the others will not. Aside from the practical problem of keeping our data synchronized across multiple storage platform, this mechanism can help us increase the availability of our data (though never reaching the unreachable 100% that some service provider boasts...). But the mechanism cannot be stretched too long: every additional cloud improves the availability a little bit, but increases our costs too: subscription costs build up linearly. How can we strengthen our protection against failures beyond the cost-availiability trade-off associated to multiple subscriptions?

A possibility is provided by insurance. Ultimately, cloud failures lead to economic losses: our time is wasted and time has an associated opportunity cost; without our data we cannot perform specific tasks, and in turn this may mean either additional costs or missed revenues. We can therefore associate a cost to the impossibility to access our data due to cloud failures. Instead of just running after failures by subscribing to multiple clouds, we can resort to cloud insurance and subscribe an insurance policy against the damage resulting from those failures. This gives us additional costs, but whose dynamics is typically different from the linear one associated to multiple cloud subscriptions. The combination of the two mechanisms may strike an optimal balance leading to a minimum cost solution. This is the gist of my recent presentation at GECON 2014. Though insurance policies are not common in economics-based service management policies, I think that they deserve to be used more often as a hedging device, instead of relying just on network-based mechanisms. You can retrieve there my paper "Balancing Leasing and Insurance Costs to Achieve Total Risk Coverage in Cloud Storage Multi-Homing".

This entry was posted in Cloud computing and tagged , , , , . Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *