2017-01-06 09:44 AM
I have been very excited about the inline data compaction feature in ONTAP 9. I read Tech ONTAP and TR4476 and learnt that inline deduplication is orthogonal to inline compaction, which means inline deduplication and online compaction can be both enabled, but a block is inline deduplicated is not compacted.
Why are this two features so mutually exclusive? Is there any nature or reason of inline dedupliation that prevents a deduplicated block from being compacted?
Any comment would be appriciated.
2017-01-06 11:59 AM
If the block is (inline) deduplicated then it's never written to disk...the metadata is simply updated to reference the block that already exists.
Hope that helps.