Big data is best known for its volume, variety, and velocity — collectively referred to as the “3 Vs” — and all three of those traits make security an elusive goal. Targeting companies grappling with that challenge, the Cloud Security Alliance on Friday released a new report offering 100 best practices.
As its name would suggest, the CSA focuses on promoting the use of security best practices within the cloud computing world; corporate members include VMware, Microsoft, AWS, and Red Hat. In an earlier report, the CSA broke down big data security risks into a set of the top 10 major challenges. Now, for each of those, it presents 10 best practices designed to help enterprises keep their information safe.
For companies working with distributed programming frameworks such as Apache Hadoop, for example, the CSA recommends using Kerberos authentication or an equivalent to help establish trust.
To ensure that the privacy of data subjects is not compromised, all personally identifiable information such as names, addresses, and Social Security numbers must be either masked or removed. It’s also important to watch for the presence of “quasi-identifiers” that can almost identify a data subject, including ZIP code, date of birth, or gender, the report warns.
Companies that use nonrelational data stores such as NoSQL databases, meanwhile, are hampered by the fact that such products typically include few robust embedded security features, the report’s authors say. For that reason, they suggest using strong encryption methods such as the Advanced Encryption Standard (AES), RSA, or Secure Hash Algorithm 2 (SHA-256) for data at rest.
“The storage of code and encryption keys must be separate from the data storage or repository,” they advise. “The encryption keys should be backed up in an offline, secured location.”
Also included in the report are suggestions for real-time security and compliance monitoring, privacy-preserving analytics, data provenance, cryptographic techniques, and more. The handbook is now available as a free download.
There’s been growing concern about the use of big data and the associated risks to privacy and security. Early this year, the U.S. Federal Trade Commission issued a report with caveats and guidelines for businesses.
Market researcher Gartner, meanwhile, predicts that the improper use of big data analytics will cause half of all business to experience ethics violations by 2018.