Amazon admins get access controls for compute and database

Need to keep devs' fingers out of production's cloud?

Amazon has brought resource-level permissions to its main compute and database services, allowing businesses to pick and choose individual staffers' level of access to specific servers or databases. The move is aimed at enterprises that need to enforce stringent identity and access management policies.

The resource-level permissions for the cloud computing firm's Elastic Compute Cloud (EC2) and Relational Database Service (RDS) update were announced by Amazon on Monday.

They build on existing identity and access technology within other services run by the compute and storage giant, such as its SSD-backed DynamoDB system, Glacier long-term storage, and SimpleDB.

Until now companies could not grant individual employees access to specific EC2 servers or RDS database instances, and could only authorize access to the entire fleet.

"This ability to assign control of an individual instance to a specific user or group helps organizations implement important security principles like separation of duties (preventing critical functions from being in the hands of one user) and least privilege (providing each user access only to the minimum resources they need to do their job)," Jim Scharf, Amazon's director of product development for its AWS Identity and Access Management division, wrote in a blog post discussing the news.

"For example, you probably don’t want to give everyone in your organization permission to terminate business-critical production instances, so now you can assign that privilege to only a few trusted administrators."

With the update, companies can use identity and access management policies to set access to EC2 servers, elastic block store (EBS) volumes, images, and elastic IP addresses. For the relational database service, IAM policies can now be applied to read replicas, along with DB Event subscriptions, option groups, parameter groups, security groups, and subnet groups.

By increasing the flexibility with which Amazon cloud customers can define and manage security policies, the company is able to offer enterprises more of the features they are familiar with when using on-premise software.

"These new features will give you more granular visibility and control over your development, test, staging and production deployments on EC2 and RDS," an Amazon spokesman wrote.

The rollout of AWS resource-level permissions for EC2 and RDS caps off a three-year journey that has seen AWS go from introducing identity and access management policies for AWS API access in 2010, to broadening support for other products, and finally to individual server access within its mainstay compute service.

Amazon has even gone as far as allowing identity verification through social networks like Facebook, but whether any sensible admin would tie AWS access to as insecure a service as an individual's personal social profile remains to be seen.

Amazon's cloud rivals, Windows Azure and Google Compute Engine, offer less detailed identity access management policies.

Microsoft's Azure Active Directory technology allows for role-based access to Azure resources, but as far as we can tell does not go down to the granular level that AWS IAM does. Google's Compute Engine is similar but lacks the on-premise integration afforded by Microsoft. ®