5–7 Oct 2026
Europe/Prague timezone

Enabling Incremental Checkpointing for GPU Workloads

7 Oct 2026, 12:00
20m
"Club A" (Prague Congress Centre)

"Club A"

Prague Congress Centre

53
Containers and checkpoint/restore MC Containers and checkpoint/restore MC

Speaker

Radostin Stoyanov (University of Oxford)

Description

With the increased adoption of AI workloads, efficient GPU checkpointing mechanisms are becoming crucial for inference, training, fine-tuning, and reinforcement learning workloads. One of the key challenges with GPU checkpointing today is the lack of memory-tracking support that enables incremental snapshots. When the GPU state is checkpointed into host memory, all pages appear modified, preventing CRIU from identifying which pages have changed since the previous checkpoint and resulting in full snapshot of the GPU memory for every iteration. In this talk, we will discuss extending the GPU plugins for CRIU with support for memory tracking that enables efficient incremental checkpointing. We will explore the benefits of this approach and the trade-offs between performance overhead and storage efficiency across different GPU workloads.

Author

Radostin Stoyanov (University of Oxford)

Presentation materials

There are no materials yet.