Forget Conventional Wisdom. Modern Encryption Technology is Ideal for Privacy Applications. Here's Why.
Your organization needs to comply with privacy regulations. Your board of directors knows the business needs to protect sensitive information as it moves among business partners, mobile users and your enterprise. Yet security technologies such as encryption are far too complex and far too difficult to deploy on a broad scale.
Actually, that is no longer the case. Leveraging identity-based encryption (IBE) is far easier and more scalable than traditional encryption technologies.
So how exactly did encryption earn its reputation for being too difficult and too costly for widespread use? Let's take a quick look at encryption's evolution, review the difficult early years and examine how today's IBE approach solves the problems from yesteryear.
In a 1995 Carnegie-Mellon University study (popularized in the paper, "Why Johnny Can't Encrypt"), sending and receiving encrypted email proved to be too hard for 75 percent of the study's participants. Fast forward 10 years and there seems to have been little progress in this area, as the title of the 2006 follow-up paper, "Why Johnny Still Can't Encrypt" indicates. If encryption isn't practical, there's no point in doing it.
The high cost of using encryption is often connected to the cost of public key infrastructure (PKI). According to the General Accounting Office, U.S. federal agencies typically spend more than $220 per digital certificate during PKI projects. In a few cases, the cost exceeded $1,000 per certificate, even exceeding $46,000 in one case. It's hard enough to do a convincing ROI calculation for many security technologies - imagine how hard it would be to justify costs like those.
Security expert Dan Geer, currently the chief scientist of Verdasys, once conjectured that the cost of using encryption is roughly the same, no matter what encryption technology you embrace. If an organization leverages symmetric encryption (technology that uses the same key to both encrypt and decrypt), Geer noted that the cost of granting the keys is high. He also noted that the cost of using asymmetric encryption (technology where one key is used to encrypt and another key to decrypt) is also high. In this case, checking keys for validity before they are used triggers most of the cost. Geer's conjecture indicates that organizations shouldn't expect to escape the high cost of encryption, regardless of our approach.
1n 2001, Professors Dan Boneh of Stanford University and Matt Franklin of the University of California, Davis, invented a new type of cryptography that violates Geer's principle. Specifically, they created the first practical and secure identity-based or IBE algorithm.
IBE leverages a user's identity for his or her key instead of using a random collection of bits. This approach eliminates many of the difficulties with traditional encryption and therefore makes encryption a more cost-effective solution for meeting today's data privacy regulations.
The benefits don't end there. IBE is simpler that PKI, and therefore has a lower total cost of ownership (TCO). In fact, IBE is more than six-times less expensive than PKI alternatives, according to Ferris Research.
Protecting the data instead of the network is now feasible, and the goal of eliminating the need to maintain a strong, well-defined network security perimeter is a realistic one.
Your Identity, Your Privacy
IBE can use almost anything as a person's identity, an email, IP or hardware address - as long as it's unique. Today, more than 5 million users worldwide leverage IBE to encrypt email messages, and for most, their email addresses are their identities.
A big benefit to organizations is that it is also quite easy to include policy information in an IBE key. Instead of encrypting using the identity -"
," it's just as easy to encrypt using the identity "
The way to calculate an IBE encryption key is publicly known, and all IBE-enabled applications can do it. Data from existing Identity and Access Management (IAM) systems can define identities, helping increase the return on that investment as IBE is used to solve the problem of managing data privacy.
For decryption to take place using IBE, a user has to be able to authenticate to a key server. The user who requests a key needs to prove he or she are authorized to receive it. If "
&classification=PCI" is used as an IBE encryption key, for example, a user might have to prove both that he or she owns the email account (
), and that he or she is entitled access to the PCI information before the key server grants the decryption key.
This ability to implement policy within the keys separates IBE from other encryption technologies. Further, IBE can easily implement the type of complex policies that data privacy regulations require.
Supporting Structured & Unstructured Data
The ease of using IBE to encrypt data is not affected by the level of structure in it. Traditionally, it has been manageable to encrypt a database, a case where the data is highly structured. Encrypting data with less structure, like email, has been more difficult, which has limited the adoption of the technology. Even more challenging has been encrypting unstructured data, like documents and spreadsheets, which can reside anywhere on a network yet still contain sensitive information.
IBE resolves these challenges. Indeed, it makes policy-based encryption very easy to implement by embedding policy in encryption keys and requiring authentication to get the corresponding decryption keys. The same key management platform can be used for all three cases, so IBE can form the basis for an enterprise-wide key management strategy that can extend easily to include new applications.
Clearly, IBE has solved many of the encryption challenges that today's enterprises face. By embedding policy directly in keys, IBE makes it easy to enforce policy. And by leveraging existing IAM infrastructure to define identities, IBE easily integrates into existing infrastructures.
At long last, encryption will be a key enabling technology that empowers us to ensure data-centric security.
Luther Martin is chief security architect at Palo Alto, CA-based Voltage Security, Inc. (www.voltage.com). He is the author of the IETF draft standards on identity-based encryption algorithms and their use in encrypted email, and is a frequent author in the areas of information security, risk management and project management. His interests include pairing-based cryptography, business applications of information security and risk management. He holds a MS degree from The Johns Hopkins University in Electrical Engineering. He can be reached at
If you want to comment on this post, you need to login.