Linux File-Systems Keeps Getting Better, But More Improvements Are Sought

[phoronix]

Phoronix: Linux File-Systems Keeps Getting Better, But More Improvements Are Sought

Linux file-systems continue getting better along with the infrastructure around it in the VFS and block code, but still there are some pain points for both users and developers around Linux storage...

http://www.phoronix.com/scan.php?page=news_item&px=LPC2018-File-System-Pain

[bitman]

> need for richer ACLs

Please no... Unixes are great because of extremely simple access control. Seriously, go try managing some access stuff in windows. It is insane.

[Britoid]

> need for richer ACLs

Please no... Unixes are great because of extremely simple access control. Seriously, go try managing some access stuff in windows. It is insane.

Linux already has advanced ACL.

[uid313]

But Windows have async I/O.
Linux does not.

[duby229]

But Windows have async I/O.
Linux does not.

Linux has real multi-threaded IO scheduler
Windows has no concept at all of IO scheduler

[cl333r]

Linux has real multi-threaded IO scheduler
Windows has no concept at all of IO scheduler

I'm just curious, what's the real advantage of having the I/O scheduler? I mean, I have a lot more issues with Linux doing I/O on the desktop than windows, because Linux's CFQ scheduler fucks up my desktop experience when doing I/O.

[duby229]

I'm just curious, what's the real advantage of having the I/O scheduler? I mean, I have a lot more issues with Linux doing I/O on the desktop than windows, because Linux's CFQ scheduler fucks up my desktop experience when doing I/O.

Sorry to say it, but CFQ wasn't really made for desktop and actually desktops should use the no-op scheduler if they are on a spinning harddrive, if you use a SSD then BLK-MQ

EDIT: I guess I should explain a bit. HDDs have integrated drive electronics whith their own logic and their own caches and so the no-op scheduler is best for them. SSD's on the other hand are basically more or less RAID-like controllers and they need more external logic and that's what BLK-MQ was designed for.

EDIT: CFQ was really made for SCSI-like controllers where their would be tons of disk accesses and latency is split between them all. So it's not made for a typical desktop where there would be minimal disk accesses but latency is important.

EDIT: Windows basically issues IO in timeslices and it puts a context switch between every time slice. And in every scenario where the process isn't ready in time for the time slice, the time slice is filled with context switches till the next time slice...

[Vistaus]

I'm just curious, what's the real advantage of having the I/O scheduler? I mean, I have a lot more issues with Linux doing I/O on the desktop than windows, because Linux's CFQ scheduler fucks up my desktop experience when doing I/O.

So just use a different scheduler. CFQ sucks. I personally use Kyber on my laptop, but BFQ is awesome as well.

[Artemis3]

HDDs have integrated drive electronics whith their own logic and their own caches and so the no-op scheduler is best for them.

Why not deadline?

[duby229]

Why not deadline?

I suppose you would have to benchmark your typical workloads and see what works best for you. I did it a few years ago and came to the conclusion no-op feels better, but my experience there was just perceptual.