| 1 |
On Friday 20 June 2014 19:48:14 Kai Krakow wrote: |
| 2 |
> microcai <microcai@×××××××××××××.org> schrieb: |
| 3 |
> > rsync is doing bunch of 4k ramdon IO when updateing portage tree, |
| 4 |
> > that will kill SSDs with much higher Write Amplification Factror. |
| 5 |
> > |
| 6 |
> > I have a 2year old SSDs that have reported Write Amplification Factor |
| 7 |
> > of 26. I think the only reason is that I put portage tree on this SSD |
| 8 |
> > to speed it up. |
| 9 |
> |
| 10 |
> Use a file system that turns random writes into sequential writes, like the |
| 11 |
> pretty newcomer f2fs. You could try using it for your rootfs but currently I |
| 12 |
> suggest just creating a separate partition for it and either mount it as |
| 13 |
> /usr/portage or symlink that dir into this directory (that way you could |
| 14 |
> use it for other purposes, too, that generate random short writes, like log |
| 15 |
> files). |
| 16 |
|
| 17 |
Well, there's a surprise! Thanks for mentioning f2fs. I've just converted my |
| 18 |
Atom box's seven partitions to it, recompiled the kernel to include it, |
| 19 |
changed the fstab entries and rebooted. It just worked. |
| 20 |
|
| 21 |
--->8 |
| 22 |
|
| 23 |
> I'd also suggest not to use the discard mount options and instead create a |
| 24 |
> cronjob that runs fstrim on the SSD devices. But YMMV. |
| 25 |
|
| 26 |
I found that fstrim can't work on f2fs file systems. I don't know whether |
| 27 |
discard works yet. |
| 28 |
|
| 29 |
Thanks again. |
| 30 |
|
| 31 |
-- |
| 32 |
Regards |
| 33 |
Peter |
Archives