1 / 6

ExtraVirt: Detecting and recovering from transient processor faults

ExtraVirt: Detecting and recovering from transient processor faults. Dominic Lucchetti, Steve Reinhardt, Peter Chen University of Michigan. Flips Happen. Similar die area + Decreasing transition energy = Increasing risk of transient failure. Multi-Processors & Virtual Machine.

ivana
Download Presentation

ExtraVirt: Detecting and recovering from transient processor faults

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. ExtraVirt: Detecting and recovering from transient processor faults Dominic Lucchetti, Steve Reinhardt, Peter Chen University of Michigan

  2. Flips Happen Similar die area + Decreasing transition energy = Increasing risk of transient failure

  3. Multi-Processors &Virtual Machine • Multi-Processor • Ensure error independence • Enable fault detection • Efficient resource sharing • Virtual Machine • No changes to OS or applications • VM replay • Synchronize replicas • Recover correct state Replica 1 Replica 2 Device Drivers Hypervisor Replication Management Layer (RML)

  4. Example: Memory Verify • Copy on write • Reduces overhead • Protects checkpoints • Merge on checkpoint • Verify correctness • Re-execute on deviation • Memory Fault Protection • ECC against RAM faults • MMU against CPU faults Checkpoint Memory Replica 1 Replica 2 Checkpoint E E E D X C C C B B B A A A Replica 3 E D C B A

  5. Status • Present • VM Replay • Beginnings of Replication Management Layer (RML) • Still much to do… • Future • Replicate the un-replicated • Handle faults in device drivers • Expanded fault model Replica 1 Replica 2 Device Drivers Hypervisor/RML

  6. Questions?

More Related