MemVerge Transparent Checkpoint Operator Certified

The Transparent Checkpoint Operator (TCO)streamlines common challenges associated with Day 2 operations in Kubernetes environments, especially for long-running Stateful applications, including AI workloads. Used in conjunction with operations such as node draining for maintenance and scheduling, it can automate the hot-starting of pods that have been killed or evicted.

For AI workloads where GPU failures can be anticipated through health checks and other metrics, the Transparent Checkpointing Operator can more gracefully allow nodes to be taken out of service, updated, rebooted, and hot restarted with the affected workloads.

The Transparent Checkpoint Operator is designed to integrate smoothly with other native Kubernetes automation facilities.

The Transparent Checkpoint operator can be used to cut compute costs for Stateful applications on the cloud by automating their hot restart after a Spot instance pre-emption.