In several articles we’ve talked about cloud security as being as good as or better than on-premises security in general. If you still aren’t convinced, you should take a look at the super-secure systems offered by some of the cloud providers.
We learned a bit about the Amazon HSM system at the recent AWS convention we attended. Fundamentally, Amazon HSM (http://aws.amazon.com/cloudhsm/) provides a physical server with extremely high security. Not only are Amazon employees blocked from accessing the server through software mechanisms (as they are for all AWS services), but they can’t even touch the server itself. Any attempt to connect to the server physically, including attempts for hard reset or removal, will result in lockdown.
One of the common use cases for HSM is for database encryption, including Oracle and Microsoft SQL Server databases. Microsoft SQL uses an encryption methodology called Transparent Data Encryption (TDE), which utilizes a master key to encrypt the data. The TDE key is highly sensitive, because it can be used to decipher the data on the database.
Given the highly risky nature of allowing anyone to reach the TDE key, some organizations choose to use HSM, which provides both hardware and software fallbacks that will cause the system to lockdown when suspicious behavior occurs. It’s important to recognize that this does not cover any legitimate uses of data, nor does it manage database security beyond the protection of the TDE Key. In several use cases, database encryption is not enough because you still want to give access to privileged users to see part of the information and to run specific SQL commands for administrative purposes. This mainly happens when DBAs or database developers want to implement/check something in the database. In those use cases, you can implement another layer of security that can give the privileged users the exact access they need while blocking them from accessing data they don’t need to see.
Register for MSSQLtips.com Live Webinar on Thursday, August 7th: Uncover Sensitive Data in your SQL Server Databases