Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

FileJournal: New journal padding mode #6856

Closed
wants to merge 2 commits into from

Conversation

XinzeChi
Copy link
Contributor

@XinzeChi XinzeChi commented Dec 8, 2015

Current, because the requirement of DirectIo, the padding mode
is adding padding data for every entry.

In fact, we fetch a couple of entries as a system write as possible
as can. So we only make sure system-write content met the requirement
of DirectIO. We adding a padding entry at the end of system-write
content(for most case, this include a couple of entries).

User could set the osd_journal_padding_ahead to decide which mode.
osd_journal_padding_ahead = true is good choice for pcie-ssd.

This work is base on #5877.
Signed-off-by: Jianpeng Ma jianpeng.ma@intel.com
Signed-off-by: Xinze Chi xinze@xsky.com

Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
@XinzeChi
Copy link
Contributor Author

XinzeChi commented Dec 9, 2015

Benchmark result:
4k randwrite, the iops could increase from 10600 -> 11000.

Intel(R) Xeon(R) CPU E5-2620 v2 @ 2.10GHz
1 SATA SSD (2 OSD )
32G Mem.

@XinzeChi
Copy link
Contributor Author

XinzeChi commented Dec 9, 2015

@liewegas , Please review it. Thanks.

Current, because the requirement of DirectIo, the padding mode
is adding padding data for every entry.

In fact, we fetch a couple of entries as a system write as possible
as can. So we only make sure system-write content met the requirement
of DirectIO. We adding a padding entry at the end of system-write
content(for most case, this include a couple of entries).

User could set the osd_journal_padding_ahead to decide which mode.
osd_journal_padding_ahead = true is good choice for pcie-ssd.

Signed-off-by: Jianpeng Ma <jianpeng.ma@intel.com>
Signed-off-by: Xinze Chi <xinze@xsky.com>
@@ -643,7 +644,11 @@ int FileJournal::_fdump(Formatter &f, bool simple)
&bl,
&seq,
&ss);
if (result != SUCCESS) {
if (result == SKIP) {
seq = pre_seq;
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

would it be simpler to make do_read_entry not clobber &seq when returning SKIP?

@ghost
Copy link

ghost commented May 3, 2016

is this still current ? If yes, it needs a rebase ;-)

@XinzeChi
Copy link
Contributor Author

XinzeChi commented May 3, 2016

We want to focus efforts on optimizations with BlueStore. Close it.

@XinzeChi XinzeChi closed this May 3, 2016
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
5 participants