Re: performance impact of various compression schemes on a zvol
- In reply to: void : "performance impact of various compression schemes on a zvol"
- Go to: [ bottom of page ] [ top of archives ] [ this month ]
Date: Tue, 17 Oct 2023 23:54:02 UTC
On Tue, Oct 17, 2023 at 11:33:27PM +0100, void wrote: > What's the perfomance penalty on zvols of compression? > The impact on the host? > > I've searched and cannot find figures. I'm not looking for exact numbers, > more rather would zstd-9 or zstd-6 be better, more suited to zvol (compression > and speed) than lz4, possibly because it's more recent? Or would no > compression at all be 'best'? With compression, it's always going to depend on your data. You might find Rich Ercolani's talk on ZFS compression at last year's OpenZFS DevSummit useful: https://openzfs.org/wiki/OpenZFS_Developer_Summit_2022 -- Brooks