Searching databases is usually done in the clear. And even if the query is encrypted, it has to be decrypted (revealing its contents) before it can be used by a search engine. What's worse is that databases themselves are stored as plaintext, available to anyone gaining access. The smarter way to handle sensitive information would be to encrypt the queries, encrypt the database and search it in its encrypted form. Impossible until now, IBM's T.J. Watson Research Center (Yorktown Heights, N.Y.) recently described a "homomorphic" encryption scheme that allows encrypted data to be searched, sorted and processed without decrypting it.
"Since public key encryption was invented, mathematicians have tried to construct a fully homomorphic encryption scheme," said Craig Gentry, a researcher at IBM. "We've found that an ideal lattice is partly homomorphic, but noise eventually drowns out the message, so what we've done is find a way to refresh it periodically so the noise becomes small again."
Using IBM's algorithm, operations of any complexity can be performed on encrypted data, with the condition that its noise level gets periodically refreshed. Since many encryption schemes are already partially homomorphic, the key to IBM's success was the refreshing scheme.
Even the industry-standard RSA encryption scheme—named after its inventors Rivest, Shamir and Adleman—is partially homomorphic, in that it allows simple multiplication of two encrypted numbers to yield their product. What Gentry and IBM have succeeded in crafting is a fully homomorphic encryption scheme where any mathematical operation can be made to work as expected. Fully homomorphic encryption schemes theoretically allow cyphertext to be manipulated as easily as plaintext, making it perfect for modern cloud computing, where your data is located remotely.
"The point is to allow others to manipulate your encrypted data without revealing it to them. For example, in cloud computing you want to store your encrypted data files out on the cloud, so that you can access it from anywhere. But you would also like to be able to search your data with some combination of keywords, then just decrypt the query results," said Gentry.
IBM's algorithm enables queries and data to remain encrypted while searching, sorting and processing, but in practice requires longer times to compute results—since encrypted operations are much more complicated. The inefficiency of the algorithm, resulting in longer execution times, is the reason IBM says homomorphic encryption is not ready for commercialization yet.
Gentry is working to improve the algorithm's efficiency. His goal is to make it easier to perform the mathematical calculations without decreasing their security level. So far he has discovered that homomorphic encryption schemes can be scaled without explosive exponential growth—a good sign—but they still need further optimization to become efficient enough to be practical.

Good morning from Los Angeles! #ibmcloud
That's it from me! Over to North America.
The data processing of Roland Garros 2012 (#RG12) rests on IBM Private Cloud http://t.co/JUaY1ItM [French Press release]
IBM Accelerates Business from Supply to Demand with New #Cloud Offerings For Smarter Commerce http://t.co/OFxknOb0 [Press Release]
How IBM #SmartCloud Foundation technology powers cloud adoption?
IBM VP @SLHebner explains here http://t.co/sSzfa0O5 [VIDEO]
IBM's Fiona Cullen will present ‘The Power of #Cloud: Driving Business Model’ On May 24 @ Utrecht, Netherlands #cloudforum2012 #ibmcloud
Blog Post: Why service providers should not ignore cloud http://t.co/ZfQyue4r via @eMarcusNet #thoughtsoncloud
Have any #cloudmoment? Share your story with us via Twitter, Instagram, YouTube, Facebook and tag it. See other stories http://t.co/J4ntsaQ5
Sign up now for IBM #SmartCloud Enterprise! No charge for select VMs (only till May 28). More Details >> http://t.co/2LEzOUZC #ibmcloud
RT @HansMoen: See this video from @IBMCloud to learn how to cut costs in building innovation in your business http://t.co/XOyJoFn6 #clou ...