Google today announced new additions to its Confidential Computing portfolio as part of its Cloud Next OnAir event: Confidential Google Kubernetes Engine (GKE) in beta mode, and new features for Confidential VMs.
Confidential GKE Nodes is the second product in Google Cloud's Confidential Computing lineup, following the July launch of Confidential VMs, which are generally available starting today.
Built on the same technology, Confidential GKE Nodes will give businesses more options for confidential workloads when they want to use Kubernetes clusters with GKE. Users will be able to configure a GKE cluster to only deploy nodes with Confidential VM capabilities, and they can keep data encrypted in memory with a node-specific key generated and managed by the AMD EPYC processor.
In addition to making Confidential VMs generally available, Google is announcing a few new capabilities added during the beta period. One of these is the inclusion of audit reports containing detailed logs about the integrity of the AMD Secure Processor Firmware responsible for key generation in Confidential VM instances.
Google is also adding new policy controls for Confidential Computing resources. Users can now use the IAM Org Policy to define access privileges for Confidential VMs; they may also disable non-Confidential VMs running in a project. More additions support this level of enforcement: Users can rely on a combination of policy constrains and firewall rules to make sure Confidential VMs only interact with other Confidential VMs, even if they VMs are in separate projects.
Read more details here.