Earlier in Introduction to case studies we introduced the Alice and Bob case studies where Alice is deploying a private government cloud and Bob is deploying a public cloud each with different security requirements. Here we dive into their particular tenant data privacy requirements. Specifically, we will look into how Alice and Bob both handle tenant data, data destruction, and data encryption.
As stated during the introduction to Alice’s case study, data protection is of an extremely high priority. She needs to ensure that a compromise of one tenant’s data does not cause loss of other tenant data. She also has strong regulator requirements that require documentation of data destruction activities. Alice does this using the following:
As stated during the introduction to Bob’s case study, tenant privacy is of an extremely high priority. In addition to the requirements and actions Bob will take to isolate tenants from one another at the infrastructure layer, Bob also needs to provide assurances for tenant data privacy. He develops an OpenStack plug-in that tracks customer data and instances by monitoring log messages and pulling both hypervisor IDs and storage image locations. The plugin can also use built-in Linux tools to ‘scrub’ the data from disk.
Bob ensures that the systems in each geographically distributed datacenter are physically secure. Where he is using a co-located facility, Bob ensures that the systems are physically separated and only accessible through logged and monitored keycard access. Additionally, each system has hardware encryption configured prior to shipping to the co-located factility. In a stand-alone datacenter, Bob maintains proper security throughout the facility by ensuring that systems have disk encryption enabled and that staff are trained in social engineering attacks.