From: Chris Friesen <chris.friesen@windriver.com>
To: linux-lvm@redhat.com
Subject: Re: [linux-lvm] cache on SSD makes system unresponsive
Date: Tue, 24 Oct 2017 17:10:27 -0600 [thread overview]
Message-ID: <59EFC863.6090109@windriver.com> (raw)
In-Reply-To: <640472762.2746512.1508882485777@mail.yahoo.com>
On 10/24/2017 04:01 PM, matthew patton wrote:
> How in the hell is the LVM cache being used at all? It has no business
> caching ANYTHING on streaming reads. Hmm, it turns out dm-cache/lvmcache
> really is retarded. It copies data to cache on first read and furthermore
> doesn't appear to detect streaming reads which have no value for caching
> purposes.
Technically it's not entirely true to say that streaming reads have no value for
caching purposes. It's conceivable to have a workload where the same file gets
read over and over, in which case it might be useful to have it cached on an SSD.
As I understand it dm-cache is using smq, which essentially uses an LRU
algorithm. So yes, it'll read the streaming data into the cache, but the
read-once/written-never data should also be the most likely to be evicted from
the cache.
For what it's worth, the Linux kernel also copies data to the page cache on
reads, which is why they introduced posix_fadvise(POSIX_FADV_DONTNEED) to allow
the application to indicate that it's done with the data and it can be dropped
from the page cache.
Chris
next prev parent reply other threads:[~2017-10-24 23:10 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <640472762.2746512.1508882485777.ref@mail.yahoo.com>
2017-10-24 22:01 ` [linux-lvm] cache on SSD makes system unresponsive matthew patton
2017-10-24 23:10 ` Chris Friesen [this message]
[not found] <1928541660.2031191.1508802005006.ref@mail.yahoo.com>
2017-10-23 23:40 ` matthew patton
2017-10-24 15:36 ` Xen
[not found] <1714773615.1945146.1508792555922.ref@mail.yahoo.com>
2017-10-23 21:02 ` matthew patton
2017-10-23 21:54 ` Xen
2017-10-24 2:51 ` John Stoffel
[not found] <1540708205.1077645.1508602122091.ref@mail.yahoo.com>
2017-10-21 16:08 ` matthew patton
[not found] <1244564108.1073508.1508601932111.ref@mail.yahoo.com>
2017-10-21 16:05 ` matthew patton
2017-10-24 18:09 ` Oleg Cherkasov
[not found] <541215543.377417.1508458336923.ref@mail.yahoo.com>
2017-10-20 0:12 ` matthew patton
2017-10-20 6:46 ` Xen
2017-10-20 9:59 ` Oleg Cherkasov
2017-10-19 17:54 Oleg Cherkasov
2017-10-19 18:13 ` Xen
2017-10-20 10:21 ` Oleg Cherkasov
2017-10-20 10:38 ` Xen
2017-10-20 11:41 ` Oleg Cherkasov
2017-10-19 18:49 ` Mike Snitzer
2017-10-20 11:07 ` Joe Thornber
2017-10-19 19:09 ` John Stoffel
2017-10-19 19:46 ` Xen
2017-10-19 21:14 ` John Stoffel
2017-10-20 6:42 ` Xen
2017-10-19 21:59 ` Oleg Cherkasov
2017-10-20 19:35 ` John Stoffel
2017-10-21 3:05 ` Mike Snitzer
2017-10-21 14:33 ` Oleg Cherkasov
2017-10-23 10:58 ` Zdenek Kabelac
2017-10-21 2:55 ` Mike Snitzer
2017-10-21 14:10 ` Oleg Cherkasov
2017-10-23 20:45 ` John Stoffel
2017-10-20 16:20 ` lejeczek
2017-10-20 16:48 ` Xen
2017-10-20 17:02 ` Bernd Eckenfels
2017-10-24 14:51 ` lejeczek
-- strict thread matches above, loose matches on Subject: below --
2017-10-19 10:05 Oleg Cherkasov
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=59EFC863.6090109@windriver.com \
--to=chris.friesen@windriver.com \
--cc=linux-lvm@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).