Nutanix ECP Elastic Dedupe Engine fingerprints data during ingest at a 16K-block granularity using a SHA-1 hash. Intel acceleration is leveraged for the SHA-1 computation which accounts for very minimal CPU overhead. Fingerprinting is only performed on data ingest and is then stored persistently as part of the written block’s metadata. Fingerprinting during data ingest is performned on data with an I/O size of 64K or greater. In cases where fingerprinting is not done during ingest (e.g. smaller I/O sizes), fingerprinting can be done as a background process.
NOTE: Initially a 4K granularity was used for fingerprinting, however Nutanix internal testing revealed that 16K granularity offers the best blend of deduplication with reduced metadata overhead. Deduplicated data is pulled into the unified cache at a 4K granularity.
Nutanix, ECP [HCI]/5.11, Pro
Dedup/Compr. Granularity, Efficiency, Data Services