Securing the Cloud

I’ve talked here before about the growing popularity, at least as  a topic for discussion, of cloud computing: the idea of moving the “heavy lifting” of applications processing and data storage to servers accessed via the Internet, leaving a lightweight, browser-based user interface on the end user’s PC.  This change has significant implications for security and privacy.

Technology Review now has an interesting interview with Whitfield Diffie, a well-known security researcher and former Chief Security Officer at Sun Microsystems, about cloud computing security.  (Diffie, together with Martin Hellman, invented the Diffie-Hellman key exchange protocol, published in 1976, one of the seminal developments in public-key cryptography.)   Diffie points out that the primary motivation for the interest in cloud computing is economic: people will use the cloud if it can do the job significantly more economically than private provisioning.  He draws an interesting analogy with transportation:

The effect of the growing dependence on cloud computing is similar to that of our dependence on public transportation, particularly air transportation, which forces us to trust organizations over which we have no control, limits what we can transport, and subjects us to rules and schedules that wouldn’t apply if we were flying our own planes.

Of course, we typically do not operate our own planes, because it would be economically unrealistic.

Diffie also points out that the significance of some security issues depends on who the user is. For many ordinary users, trusting Google (for example) to deliver their E-mail may not be that different from trusting the US Postal Service to deliver their “snail mail”.   Some users may want or need more in terms of security: for communications applications, they can encrypt all their messages.

If you want to do processing in the cloud, then encryption alone will not do the job (although there are some interesting theoretical results showing that computation and search on encrypted data are possible in principle). You can encrypt your data as it passes to and from the cloud provider, but you must ultimately decide who you can trust. Diffie also points out that there may be a market for cloud services that offer more security:

Much of this would result from care on the part of cloud computing providers–choosing more secure operating systems such as Open BSD and Solaris–and keeping those systems carefully configured. A security-conscious computing services provider would provision each user with its own processors, caches, and memory at any given moment and would clean house between users, reloading the operating system and zeroing all memory.

And he notes, once again, that complexity is the enemy of security.