AWS Security Blog
Updated data classification whitepaper, includes guidance on establishing data classification programs
Note from March 25, 2020: This post was originally published on June 20, 2018, along with the Secure Cloud Adoption: Data Classification whitepaper, which has now been refreshed with newer information. We’ve updated our post accordingly.
We’re sharing an update to the Secure Cloud Adoption: Data Classification whitepaper to help customers address data classification. Data classification continues to be a foundational step in cybersecurity governance and risk management. It has been used for decades to help organizations safeguard sensitive or critical data with appropriate levels of protection. Regardless of whether data is processed or stored in traditional, on-premises systems or the cloud, data classification is a starting point for maintaining the confidentiality—and potentially the integrity and availability—of data based on the data’s risk impact level.
This whitepaper offers leading practices for developing a data classification strategy. The updates share how you can classify data, privacy considerations, and examples of different data classification levels and tiers of data that exist in both public and private sector enterprises. It also provides best practices—such as not over classifying data to maximize flexibility—and models you can use to classify data so that you can quickly move workloads to the cloud.
Download and read the updated whitepaper
If you have questions or want to learn more, contact your account executive or contact AWS Support.
Want more AWS Security how-to content, news, and feature announcements? Follow us on Twitter.