io-shepherding · research

I/O shepherding

I/O shepherding centralizes filesystem fault-detection and recovery policies around block requests and responses.

CrookFS, an ext3-based research prototype, selects policies by block type and volume. Policies compose retries, checksums, replicas or repair. [1]

Recovery can create persistent state: relocating a failed checkpoint write also changes allocation and mapping metadata. Chained transactions commit those changes before releasing the original transaction. [1]

Relations

related
Checksum, Intra-disk redundancy, Lost write

Limitations

  • Journal space must accommodate chained work; protection of the journal itself needs restricted policies. [1]

Citations

  1. Gunawi et al. Improving File System Reliability with I/O Shepherding. SOSP 2007, sections 3-5. (accessed 2026-11-09)

Further reading

  1. Hesela: recovery policies, checkpoint repair and memory boundaries (accessed 2026-11-09)
Sources & caveats
  • Historical Linux 2.6.12 prototype and selected fault-injection workloads; not a feature claim about current ext3 or ext4. [1]

JSON: /data/terms/io-shepherding.json

Published by Hesela team · Updated

Report a correction · Editorial standards · Reuse guidance

Cite this page
Hesela team. I/O shepherding. Hesela. 2026-11-09. https://hesela.dev/terms/io-shepherding/