2013-7-31， 2:01， Sage Weil s...@inktank.com wrote:
Hi Haomai,
On Wed, 31 Jul 2013, Haomai Wang wrote:
Every node of ceph cluster has a backend filesystem such as btrfs,
xfs and ext4 that provides storage for data objects, whose location
are determined by CRUSH algorithm. There should
On Aug 28, 2013, at 7:01 AM, Sage Weil s...@inktank.com wrote:
Hi Haomai,
I just wanted to check in to see if things have progressed at all since we
talked at CDS. If you have any questions or there is anything I can
help with, let me know! I'd love to see this alternative backend make
Hi all,
Now OpenStack Nova master branch still exists a bug when you boot a VM which
root disk size is specified. The storage backend of Nova also is rbd. For
example, you boot a VM and specify 10G as root disk size. But the image is only
1G. Then VM will be spawned and the root disk size will
On Nov 13, 2013, at 10:58 AM, Josh Durgin josh.dur...@inktank.com wrote:
On 11/11/2013 11:10 PM, Haomai Wang wrote:
Hi all,
Now OpenStack Nova master branch still exists a bug when you boot a VM which
root disk size is specified. The storage backend of Nova also is rbd. For
example, you
On Nov 13, 2013, at 9:14 PM, Haomai Wang haomaiw...@gmail.com wrote:
On Nov 13, 2013, at 10:58 AM, Josh Durgin josh.dur...@inktank.com wrote:
On 11/11/2013 11:10 PM, Haomai Wang wrote:
Hi all,
Now OpenStack Nova master branch still exists a bug when you boot a VM
which root disk size
Yes, we have related bp for rocksdb backend.
wiki.ceph.com/index.php?title=01Planning/02Blueprints/Firefly/osd:_new_key%2F%2Fvalue_backend
On Wed, Nov 27, 2013 at 6:56 PM, Stefan Priebe - Profihost AG
s.pri...@profihost.ag wrote:
the performance comparisions are very impressive:
On Dec 13, 2013, at 1:01 AM, Sage Weil s...@inktank.com wrote:
On Thu, 12 Dec 2013, Haomai Wang wrote:
On Thu, Dec 12, 2013 at 1:26 PM, Sage Weil s...@inktank.com wrote:
[adding cc ceph-devel]
[attempt 2]
On Wed, 11 Dec 2013, Haomai Wang wrote:
Hi Sage,
Since last CDS, you have
On Dec 22, 2013, at 1:20 PM, Sage Weil s...@inktank.com wrote:
On Sat, 21 Dec 2013, Haomai Wang wrote:
On Dec 13, 2013, at 1:01 AM, Sage Weil s...@inktank.com wrote:
On Thu, 12 Dec 2013, Haomai Wang wrote:
On Thu, Dec 12, 2013 at 1:26 PM, Sage Weil s...@inktank.com wrote:
[adding cc ceph
On Dec 22, 2013, at 2:02 PM, Haomai Wang haomaiw...@gmail.com wrote:
On Dec 22, 2013, at 1:20 PM, Sage Weil s...@inktank.com wrote:
On Sat, 21 Dec 2013, Haomai Wang wrote:
On Dec 13, 2013, at 1:01 AM, Sage Weil s...@inktank.com wrote:
On Thu, 12 Dec 2013, Haomai Wang wrote:
On Thu
Hi all,
We know FileJournal plays a important role in FileStore backend, it can
hugely reduce write latency and improve small write operations.
But in practice, there exists exceptions such as we already use FlashCache or
cachepool(although it's not ready).
If cachepool enabled, we may use use
on
FileStore to implement it.
-Greg
Software Engineer #42 @ http://inktank.com | http://ceph.com
On Thu, Jan 9, 2014 at 12:13 AM, Haomai Wang haomaiw...@gmail.com wrote:
Hi all,
We know FileJournal plays a important role in FileStore backend, it can
hugely reduce write latency and improve
if the FileJournal could be disabled, there must be
something else to implement the Transaction Interface. But it seems
hard while no local file-system provide such function in my opinion.
On 10 January 2014 10:04, Haomai Wang haomaiw...@gmail.com wrote:
On Fri, Jan 10, 2014 at 1:28 AM, Gregory
, Haomai Wang haomaiw...@gmail.com wrote:
On Fri, Jan 10, 2014 at 11:13 AM, Gregory Farnum g...@inktank.com wrote:
Exactly. We can't do a safe update without a journal — what if power
goes out while the write is happening? When we boot back up, we don't
know what version the object is actually
It's great to performance, more advice from client is welcomed for ObjectStore
implementation.
Maybe the new operation can be just like fadvise, accept flags as argument and
ObjectStore will try to do it but not certainly.
KeyValueStore will also enjoy it and get much benefit from it.
On Sat,
will expose header to caller among write operations.
StripObjectMap is the successor of GenericObjectMap and used by KeyValueStore
directly. It encapsulate the interface of GenericObjectMap and make more
suitable for KeyValueStore.
Best regards,
Haomai Wang, UnitedStack Inc.
--
To unsubscribe from
Hi all,
I noticed a pull request(https://github.com/ceph/ceph/pull/1152) which
make a deal of optimizations on FileStore. Although the way modify ObjecStore
and caller's codes are not accepted for me, it still seemed good for
performance purpose.
I agree with we need a session mechanism to cache
Hi all,
last release I propose a KeyValueStore prototype(get info from
http://sebastien-han.fr/blog/2013/12/02/ceph-performance-interesting-things-going-on).
It contains some performance results and problems. Now I'd like to
refresh our thoughts on KeyValueStore.
KeyValueStore is pursuing
On Sat, Mar 1, 2014 at 8:04 AM, Danny Al-Gaaf danny.al-g...@bisect.de wrote:
Hi,
Am 28.02.2014 03:45, schrieb Haomai Wang:
[...]
I use fio which rbd supported from
TelekomCloud(https://github.com/TelekomCloud/fio/commits/rbd-engine)
to test rbd.
I would recommend to no longer use
I think the reason why the little difference between leveldb and
rocksdb in FileStore is that the main latency cause isn't KeyValueDB
backend.
So we may not get enough benefit from rocksdb instead of leveldb by FileStore.
On Wed, Mar 5, 2014 at 4:23 PM, Alexandre DERUMIER aderum...@odiso.com
Hi all,
As we know, snapshot is a lightweight resource in librbd and we
doesn't have any statistic informations about it. But it causes some
problems to the cloud management.
We can't measure the size of snapshot, different snapshot will occur
different space. So we don't have way to estimate
Hi all,
I would like to share some ideas about how to improve performance on
ceph with SSD. Not much preciseness.
Our ssd is 500GB and each OSD own a SSD(journal is on the same SSD).
ceph version is 0.67.5(Dumping)
At first, we find three bottleneck on filestore:
1. fdcache_lock(changed in
the new coming leveldb backend store help for this specific case ?
- Mail original -
De: Gregory Farnum g...@inktank.com
À: Haomai Wang haomaiw...@gmail.com
Cc: ceph-devel@vger.kernel.org
Envoyé: Mercredi 9 Avril 2014 16:15:14
Objet: Re: [Share]Performance tunning on Ceph FileStore
So cool!
I remember a telekomcloud guy is also worked for it since last summit.
On Wed, Apr 30, 2014 at 6:17 AM, Samuel Just sam.j...@inktank.com wrote:
I'm starting on adding systemtap trace points to the osd.
https://github.com/athanatos/ceph/tree/wip-osd-stap
has some initial build
Congratulation!
On Wed, Apr 30, 2014 at 8:18 PM, Sage Weil s...@inktank.com wrote:
Today we are announcing some very big news: Red Hat is acquiring Inktank.
We are very excited about what this means for Ceph, the community, the
team, our partners, and our customers. Ceph has come a long way in
Hi all,
I remember there exists discuss about DB(mysql) performance on rbd.
Recently I test mysql-bench with rbd and found awful performance. So I
dive into it and find that main cause is flush request from guest.
As we know, applications such as mysql, ceph has own journal for
durable and
the root cause for db on rbd performance.
On Wed, May 21, 2014 at 6:15 PM, Haomai Wang haomaiw...@gmail.com wrote:
Hi all,
I remember there exists discuss about DB(mysql) performance on rbd.
Recently I test mysql-bench with rbd and found awful performance. So I
dive into it and find that main
-ow...@vger.kernel.org
[mailto:ceph-devel-ow...@vger.kernel.org] On Behalf Of Haomai Wang
Sent: Wednesday, 21 May, 2014 6:22 PM
To: ceph-devel@vger.kernel.org
Subject: Re: [Performance] Improvement on DB Performance
I pushed the commit to fix this
problem(https://github.com/ceph/ceph/pull
On Mon, May 26, 2014 at 9:46 AM, Guang Yang yguan...@outlook.com wrote:
Hello Haomai,
We are evaluating the key-value store backend which comes along with Firefly
release (thanks for implementing it in Ceph), it is very promising for a
couple of our use cases, after going through the related
On Mon, May 26, 2014 at 5:23 PM, Wido den Hollander w...@42on.com wrote:
On 05/26/2014 06:55 AM, Haomai Wang wrote:
On Mon, May 26, 2014 at 9:46 AM, Guang Yang yguan...@outlook.com wrote:
Hello Haomai,
We are evaluating the key-value store backend which comes along with
Firefly release
On Wed, May 21, 2014 at 11:23 PM, Sage Weil s...@inktank.com wrote:
On Wed, 21 May 2014, Haomai Wang wrote:
I pushed the commit to fix this
problem(https://github.com/ceph/ceph/pull/1848).
With test program(Each sync request is issued with ten write request),
a significant improvement
performance improvements with this branch.
Greets,
Stefan
Am 09.04.2014 12:05, schrieb Haomai Wang:
Hi all,
I would like to share some ideas about how to improve performance on
ceph with SSD. Not much preciseness.
Our ssd is 500GB and each OSD own a SSD(journal is on the same SSD).
ceph
I'm not full sure the correctness of changes although it seemed ok to
me. And I apply these changes to product env and no problems.
On Tue, May 27, 2014 at 2:05 PM, Stefan Priebe - Profihost AG
s.pri...@profihost.ag wrote:
Am 27.05.2014 06:42, schrieb Haomai Wang:
On Tue, May 27, 2014 at 4:29
Still not, I will try to push to master branch
On Tue, May 27, 2014 at 2:45 PM, Stefan Priebe - Profihost AG
s.pri...@profihost.ag wrote:
Am 27.05.2014 08:37, schrieb Haomai Wang:
I'm not full sure the correctness of changes although it seemed ok to
me. And I apply these changes to product env
Hi Gregory,
I try to reproduce the bug in my local machine but failed.
My test cmdline:
./ceph_test_rados --op read 100 --op write 100 --op delete 50
--max-ops 40 --objects 1024 --max-in-flight 64 --size 400
--min-stride-size 40 --max-stride-size 80 --max-seconds 600
--op
using similar conf file as yours. Is
this the expected behavior or am I missing something?
Thanks,
Sushma
On Fri, Feb 28, 2014 at 11:00 PM, Haomai Wang haomaiw...@gmail.com wrote:
On Sat, Mar 1, 2014 at 8:04 AM, Danny Al-Gaaf danny.al-g...@bisect.de
wrote:
Hi,
Am 28.02.2014 03:45
.
On Sat, May 31, 2014 at 2:06 AM, Gregory Farnum g...@inktank.com wrote:
On Fri, May 30, 2014 at 2:18 AM, Haomai Wang haomaiw...@gmail.com wrote:
Hi Gregory,
I try to reproduce the bug in my local machine but failed.
My test cmdline:
./ceph_test_rados --op read 100 --op write 100 --op delete 50
/familiar/not familiar/
On Tue, Jun 3, 2014 at 10:33 PM, Haomai Wang haomaiw...@gmail.com wrote:
Hi Gregory,
I checked again and again each line change about spill out codes,
still failed to find anything wrong.
I ran ceph_test_rados then activate scrub process several times
locally
The fix pull request is https://github.com/ceph/ceph/pull/1912/files.
Someone can help to review and merge
On Wed, Jun 4, 2014 at 3:38 AM, Sushma R gsus...@gmail.com wrote:
ceph version : master (ceph version 0.80-713-g86754cc
(86754cc78ca570f19f5a68fb634d613f952a22eb))
fio version :
Hi,
Previously I sent a mail about the difficult of rbd snapshot size
statistic. The main solution is using object map to store the changes.
The problem is we can't handle with multi client concurrent modify.
Lack of object map(like pointer map in qcow2), it cause many problems
in librbd. Such as
On Thu, Jun 5, 2014 at 3:25 PM, Wido den Hollander w...@42on.com wrote:
On 06/05/2014 09:01 AM, Haomai Wang wrote:
Hi,
Previously I sent a mail about the difficult of rbd snapshot size
statistic. The main solution is using object map to store the changes.
The problem is we can't handle
[mailto:ceph-devel-ow...@vger.kernel.org] On Behalf Of Haomai Wang
Sent: Thursday, June 05, 2014 12:43 AM
To: Wido den Hollander
Cc: Sage Weil; Josh Durgin; ceph-devel@vger.kernel.org
Subject: Re: [Feature]Proposal for adding a new flag named shared to support
performance and statistic
, Jun 3, 2014 at 10:33 PM, Haomai Wang haomaiw...@gmail.com wrote:
/familiar/not familiar/
On Tue, Jun 3, 2014 at 10:33 PM, Haomai Wang haomaiw...@gmail.com wrote:
Hi Gregory,
I checked again and again each line change about spill out codes,
still failed to find anything wrong.
I ran
The fix should make clone method copy cephos prefix xattrs
On Sat, Jun 7, 2014 at 2:54 AM, Haomai Wang haomaiw...@gmail.com wrote:
Hi Greg,
I have found the reason. user.cephos.spill_out can't be apply to new
object when calling clone method. So if the origin object is spill
out, the new
Yes, maybe you can add it in your branch. Because it will happen when
creating object and set spill out
On Sat, Jun 7, 2014 at 2:59 AM, Gregory Farnum g...@inktank.com wrote:
On Fri, Jun 6, 2014 at 11:55 AM, Haomai Wang haomaiw...@gmail.com wrote:
The fix should make clone method copy cephos
Greg, I have submit a patch to fix it. https://github.com/ceph/ceph/pull/1932
On Sat, Jun 7, 2014 at 3:00 AM, Haomai Wang haomaiw...@gmail.com wrote:
Yes, maybe you can add it in your branch. Because it will happen when
creating object and set spill out
On Sat, Jun 7, 2014 at 2:59 AM, Gregory
bit :P. I'm
focusing on instrumenting the OSD op life cycle.
-Sam
On Wed, Apr 30, 2014 at 6:19 AM, Danny Al-Gaaf danny.al-g...@bisect.de
wrote:
Am 30.04.2014 04:21, schrieb Haomai Wang:
So cool!
I remember a telekomcloud guy is also worked for it since last summit.
Correct ... our
Thanks, Josh!
Your points are really helpful. Maybe we can schedule this bp to the
near CDS? The implementation I hope can has great performance effects
on librbd.
On Tue, Jun 10, 2014 at 9:16 AM, Josh Durgin josh.dur...@inktank.com wrote:
On 06/05/2014 12:01 AM, Haomai Wang wrote:
Hi
Hi,
Now librbd is the only user for set_alloc_hint op.
void AioWrite::add_write_ops(librados::ObjectWriteOperation wr) {
wr.set_alloc_hint(m_ictx-get_object_size(), m_ictx-get_object_size());
wr.write(m_object_off, m_write_data);
}
According above, the arguments for set_alloc_hint
I would like to say that MemStore isn't a good backend for evaluating
performance, it's just a prototype for ObjectStore.
On Tue, Jun 24, 2014 at 8:13 PM, Andreas Joachim Peters
andreas.joachim.pet...@cern.ch wrote:
I made the same MemStore measurements with the master branch.
It seems that the
As I mentioned days ago:
There exists two points related kvstore perf:
1. The order of image and the strip
size are important to performance. Because the header like inode in fs is
much lightweight than fd, so the order of image is expected to be
lower. And strip size can be configurated to 4kb
you need to pay attention to fiemap option, ceph --show-config | grep fiemap
There exists some tricks about fiemap, please search it at tracker.ceph.com
On Fri, Jun 27, 2014 at 12:27 AM, huang jun hjwsm1...@gmail.com wrote:
hi,all
We migrate vmware EXSI vm from ceph cluster via NFS to local
understand the reason for this lock and whether it can be
replaced with a RWLock or any other suggestions to avoid serialization due to
this lock?
Thanks,
Sushma
-Original Message-
From: Haomai Wang [mailto:haomaiw...@gmail.com]
Sent: Friday, June 27, 2014 1:08 AM
To: Sushma
Of Haomai Wang
Sent: Monday, June 30, 2014 11:10 PM
To: Sushma Gurram
Cc: Shu, Xinxin; Mark Nelson; Sage Weil; Zhang, Jian;
ceph-devel@vger.kernel.org
Subject: Re: [RFC] add rocksdb support
Hi Sushma,
Thanks for your investigations! We already noticed the serializing risk on
GenericObjectMap
DBObjectMap header cache, if enable header cache,
is the header_lock still be a awful point? Same as KeyValueStore, I
will try to see too.
Thanks,
Sushma
-Original Message-
From: Haomai Wang [mailto:haomaiw...@gmail.com]
Sent: Tuesday, July 01, 2014 1:06 AM
To: Somnath Roy
Cc: Sushma
implementation).
Thanks,
Sushma
-Original Message-
From: Haomai Wang [mailto:haomaiw...@gmail.com]
Sent: Tuesday, July 01, 2014 10:03 AM
To: Sushma Gurram
Cc: Somnath Roy; Shu, Xinxin; Mark Nelson; Sage Weil; Zhang, Jian;
ceph-devel@vger.kernel.org
Subject: Re: [RFC] add rocksdb
Could you give some perf counter from rbd client side? Such as op latency?
On Wed, Jul 2, 2014 at 9:01 PM, Stefan Priebe - Profihost AG
s.pri...@profihost.ag wrote:
Am 02.07.2014 00:51, schrieb Gregory Farnum:
On Thu, Jun 26, 2014 at 11:49 PM, Stefan Priebe - Profihost AG
s.pri...@profihost.ag
exclusive lock
, there maybe some unsafe scenarios . I am not sure whether my understanding
is right ? if my understanding is right , I think RWlock or a fine-grain
lock is a good suggestion.
-Original Message-
From: Haomai Wang [mailto:haomaiw...@gmail.com]
Sent: Tuesday, July 01, 2014
down anywhere, Josh?
-Greg
On Tue, Jun 10, 2014 at 12:38 PM, Josh Durgin josh.dur...@inktank.com wrote:
On Tue, 10 Jun 2014 14:52:54 +0800
Haomai Wang haomaiw...@gmail.com wrote:
Thanks, Josh!
Your points are really helpful. Maybe we can schedule this bp to the
near CDS? The implementation I
Hi Kasper,
Can you talk about how you make use of Ceph and detail information on CentOS?
I guess that you use the CephFS on Ceph Cluster?
Best regards,
Wheats
在 2013-7-17，下午2:16，Kasper Dieter dieter.kas...@ts.fujitsu.com 写道：
Hi Xiaoxi,
we are really running Ceph on CentOS-6.4
(6 server
Every node of ceph cluster has a backend filesystem such as btrfs,
xfs and ext4 that provides storage for data objects, whose location
are determined by CRUSH algorithm. There should exists an abstract
interface sitting between osd and backend store, allowing different
backend store
Thanks, I will dive into it and fix it next.
On Tue, Jul 22, 2014 at 11:49 PM, Sage Weil sw...@redhat.com wrote:
Hi Haomai,
Hmm, one other thing: I'm testing the fix in wip-8701 and it is tripping
over the KeyValueStore test. This
./ceph_test_objectstore
Hi sage,
The fix is https://github.com/ceph/ceph/pull/2136. :-)
On Wed, Jul 23, 2014 at 12:48 AM, Haomai Wang haomaiw...@gmail.com wrote:
Thanks, I will dive into it and fix it next.
On Tue, Jul 22, 2014 at 11:49 PM, Sage Weil sw...@redhat.com wrote:
Hi Haomai,
Hmm, one other thing: I'm
Awesome job!
On Thu, Jul 31, 2014 at 1:49 PM, Mark Kirkwood
mark.kirkw...@catalyst.net.nz wrote:
On 31/07/14 17:25, Sage Weil wrote:
After the latest set of bug fixes to the FileStore file naming code I am
newly inspired to replace it with something less complex. Right now I'm
mostly
Hi Sage, Josh:
ImageIndex is aimed to hold each object's location info which avoid
extra checking for none-existing object. It's only used when image flags
exists LIBRBD_CREATE_NONSHARED. Otherwise, ImageIndex will become gawp and
has no effect.
Each object has three state:
1. UNKNOWN: default
On Wed, Aug 13, 2014 at 10:14 AM, Josh Durgin josh.dur...@inktank.com wrote:
On 08/11/2014 07:50 PM, Haomai Wang wrote:
Hi Sage, Josh:
ImageIndex is aimed to hold each object's location info which avoid
extra checking for none-existing object. It's only used when image flags
exists
, 2014 7:45 AM
To: Haomai Wang; Sage Weil
Cc: ceph-devel@vger.kernel.org
Subject: Re: [RFC]About ImageIndex
On 08/11/2014 07:50 PM, Haomai Wang wrote:
Hi Sage, Josh:
ImageIndex is aimed to hold each object's location info which avoid
extra checking for none-existing object. It's only used
Hi,
Could you give some more detail infos such as operation before occur errors?
And what's your ceph version?
On Fri, Sep 5, 2014 at 3:16 PM, 廖建锋 de...@f-club.cn wrote:
Dear CEPH ,
Urgent question, I met a FAILED assert(0 == unexpected error)
yesterday , Now i have not way to start
?
发件人: Somnath Roy [somnath@sandisk.com]
发送时间: 2014年9月7日 1:12
到: Haomai Wang; 廖建锋
Cc: ceph-users; ceph-devel
主题: RE: [ceph-users] ceph osd unexpected error
Have you set the open file descriptor limit in the OSD node ?
Try setting it like 'ulimit -n 65536
-Original Message-
From
tell more about your
key/value interface. I'm doing some jobs for NVMe interface with intel
NVMe SSD.
Thanks Regards
Somnath
-Original Message-
From: Sage Weil [mailto:sw...@redhat.com]
Sent: Thursday, September 11, 2014 6:31 PM
To: Somnath Roy
Cc: Haomai Wang (haomaiw
Hi all,
Recently, I did some basic work on new message implementation based on
event(https://github.com/yuyuyu101/ceph/tree/msg-event). The basic
idea is that we use a Processor thread for each Messenger to monitor
all sockets and dispatch fd to threadpool. The event mechanism can be
epoll,
at 11:51 PM, Sage Weil sw...@redhat.com wrote:
Hi Haomai,
On Fri, 12 Sep 2014, Haomai Wang wrote:
Hi all,
Recently, I did some basic work on new message implementation based on
event(https://github.com/yuyuyu101/ceph/tree/msg-event). The basic
idea is that we use a Processor thread for each
According http://tracker.ceph.com/issues/9513, do you mean that rbd
cache will make 10x performance degradation for random read?
On Thu, Sep 18, 2014 at 7:44 AM, Somnath Roy somnath@sandisk.com wrote:
Josh/Sage,
I should mention that even after turning off rbd cache I am getting ~20%
Good point, but do you have considered that the impaction for write
ops? And if skip page cache, FileStore is responsible for data cache?
On Wed, Sep 24, 2014 at 3:29 AM, Sage Weil sw...@redhat.com wrote:
On Tue, 23 Sep 2014, Somnath Roy wrote:
Milosz,
Thanks for the response. I will see if I
.
Thanks Regards
Somnath
-Original Message-
From: Haomai Wang [mailto:haomaiw...@gmail.com]
Sent: Tuesday, September 23, 2014 7:07 PM
To: Sage Weil
Cc: Somnath Roy; Milosz Tanski; ceph-devel@vger.kernel.org
Subject: Re: Impact of page cache on OSD read performance for SSD
Good
On Wed, Sep 24, 2014 at 8:38 PM, Sage Weil sw...@redhat.com wrote:
On Wed, 24 Sep 2014, Haomai Wang wrote:
I agree with that direct read will help for disk read. But if read data
is hot and small enough to fit in memory, page cache is a good place to
hold data cache. If discard page cache, we
direct_io read option
(Need to quantify direct_io write also) as Sage suggested.
Thanks Regards
Somnath
-Original Message-
From: Haomai Wang [mailto:haomaiw...@gmail.com]
Sent: Wednesday, September 24, 2014 9:06 AM
To: Sage Weil
Cc: Somnath Roy; Milosz Tanski; ceph-devel
)
On Fri, Sep 26, 2014 at 10:27 AM, Haomai Wang haomaiw...@gmail.com wrote:
Thanks for sage!
I'm on the flight at Oct 1. :-(
Now my team is mainly worked on the performance of ceph, we have
observed these points:
1. encode/decode plays remarkable latency, especially in
ObjectStore::Transaction
No problem, I would like do it.
On Tue, Sep 30, 2014 at 7:48 AM, Sage Weil sw...@redhat.com wrote:
Hi Haomai,
Not sure if you saw
http://tracker.ceph.com/issues/9580
which came from an issue Mark had getting ceph-disk to work with the
keyvaluestore backend. I think the answer is
Hi sage,
What do you think use existing ObjectStore::peek_journal_fsid
interface to detect whether journal needed.
KeyValueStore and MemStore could set passing argument fsid to zero
to indicate no journal.
On Tue, Sep 30, 2014 at 10:23 AM, Haomai Wang haomaiw...@gmail.com wrote:
No problem, I
Thanks for Mark!
It's a pity that I can't join it while on the flight. I hope I have
time to view the video(exists?).
As a reminder,
AsyncMessenger(https://github.com/yuyuyu101/ceph/tree/msg-event-worker-mode)
is ready to test for developer. For io depth 1(4k randwrite),
AsyncMessenger saves
On Thu, Oct 2, 2014 at 7:53 AM, Sage Weil sw...@redhat.com wrote:
On Thu, 2 Oct 2014, Mark Kirkwood wrote:
On 30/09/14 17:05, Sage Weil wrote:
On Tue, 30 Sep 2014, Haomai Wang wrote:
Hi sage,
What do you think use existing ObjectStore::peek_journal_fsid
interface to detect whether
Correctly, maybe we can move these super metadata to backend!
On Fri, Oct 3, 2014 at 6:47 AM, Somnath Roy somnath@sandisk.com wrote:
Hi Sage/Haomai,
I was going through the key/value store implementation and have one basic
question regarding the way it is designed.
I think key/value
Thanks Somnath!
I have another simple performance test for async messenger:
For 4k object read, master branch used 4.46s to complete tests, async
Messenger used 3.14s
For 4k object write, master branch used 10.6s to complete, async
Messenger used 6.6s!!
Detailf results see below, 4k object read
[cc to ceph-devel]
On Tue, Oct 21, 2014 at 11:51 PM, Sage Weil s...@newdream.net wrote:
Hi Haomai,
You and your team have been doing great work and I'm very happy that you
are working with Ceph! The performance gains you've seen are very
encouraging.
1. Use AsyncMessenger for both
? OR I need to add some config
option for that ?
3. Other than ms_event_op_threads , is there any tunable parameter I should
be playing with ?
Thanks Regards
Somnath
-Original Message-
From: Haomai Wang [mailto:haomaiw...@gmail.com]
Sent: Saturday, October 18, 2014 10:15 PM
with the client which is not using this async
messenger for example krbd ?
Regards
Somnath
-Original Message-
From: Haomai Wang [mailto:haomaiw...@gmail.com]
Sent: Wednesday, October 22, 2014 10:02 AM
To: Somnath Roy
Cc: ceph-devel@vger.kernel.org
Subject: Re: The Async messenger benchmark
https://wiki.ceph.com/Planning/Blueprints/Submissions/Fixed_memory_layout_for_Message%2F%2FOp_passing
@sage, thanks
On Thu, Oct 23, 2014 at 12:24 AM, Sage Weil s...@newdream.net wrote:
Hi Everyone,
Just a reminder that there won't be a performance call next week because
of CDS, which is
I'm interested in bufferlist's own encode/decode performance. But as I
performed until now, I think we need to consider change caller's
behavior to get better performance.
Combined
(https://wiki.ceph.com/Planning/Blueprints/Hammer/Fixed_memory_layout_for_Message%2F%2FOp_passing)
with
Thanks for Loic!
I will join.
On Thu, Oct 30, 2014 at 1:54 AM, Loic Dachary l...@dachary.org wrote:
Hi Ceph,
TL;DR: Register for the Micro Ceph and OpenStack Design Summit November 3rd,
2014 11:40am
http://kilodesignsummit.sched.org/event/f2e49f4547a757cc3d51f5641b2000cb
November 3rd,
Could you describe more about 2x7 iops?
So you mean 8 OSD each backend with SSD can achieve with 14w iops?
is it read or write? could you give fio options?
On Fri, Oct 31, 2014 at 12:01 AM, Alexandre DERUMIER
aderum...@odiso.com wrote:
I'll try to add more OSD next week, if it's scale it's a
an instance of 'ceph::FailedAssertion'
Abandon
- Mail original -
De: Haomai Wang haomaiw...@gmail.com
À: Alexandre DERUMIER aderum...@odiso.com
Cc: Somnath Roy somnath@sandisk.com, ceph-devel@vger.kernel.org,
Sage Weil s...@newdream.net
Envoyé: Mardi 4 Novembre 2014 10:29:46
Objet: Re
Hi Chendi,
It seemed that you find two bugs for KeyValueStore:
1. Potential race conflict with strip_header-buffers:
strip_header is owner by a thread who want to access header. Now
KeyValueStore in order to avoid lock bottleneck, only use
Sequencer-level(much like PG) to solve concurrent
On Sun, Nov 9, 2014 at 5:59 AM, Sage Weil s...@newdream.net wrote:
On Sat, 8 Nov 2014, Haomai Wang wrote:
As for OOM, I think the root cause is the mistake commit above too.
Because meta collection will be updated each transaction and
StripObjectHeader::buffers will be always kept in memory
On Wed, Nov 12, 2014 at 9:51 PM, Sage Weil sw...@redhat.com wrote:
On Wed, 12 Nov 2014, xinxin shu wrote:
recently we focus on 4k random write , dump transaction on every 4k
random write op , found that , for every 4k random write , it will
update pg epoch and pg info on OSDService::infos_oid
Hmm, I think it's a good perf topic to discuss about buffer
alloc/dealloc. For example, maybe frequency alloced object can use
memory pool(each pool stores the same objects), but the most challenge
to this is also STL structures.
On Fri, Nov 14, 2014 at 1:05 AM, Mark Nelson
It's surprised that the test machine run on a very new kernel but
occur this problem:
plana 47 is 12.04.5 with kernel 3.18.0-rc6-ceph-00024-geb0e5fd
plana 50 is 12.04.4 with kernel 3.17.0-rc6-ceph-2-ge8acad6
Which local filesystem is ran on?
On Tue, Nov 25, 2014 at 5:03 AM, Samuel Just
Oh, sorry. This series seemed that already backported...
On Tue, Nov 25, 2014 at 10:14 AM, Haomai Wang haomaiw...@gmail.com wrote:
Backport this
series(https://github.com/yuandong1222/ceph/commit/3f8fb85b341726ae7bb44b4c699707333cd63ccd)
and test again ? Or
Actually, my up to ten ceph
think the bug is actually with FileStore::_do_sparse_copy_range,
which isn't truncating out properly to create 0 filled sections.
-Sam
On Mon, Nov 24, 2014 at 6:17 PM, Haomai Wang haomaiw...@gmail.com wrote:
Oh, sorry. This series seemed that already backported...
On Tue, Nov 25, 2014 at 10:14
Hi Gregoy and Sage,
I'm just writing Messenger's unit tests to ensure that SimpleMessenger
and AsyncMessenger behavior same and expected.
I think the most unclear thing is the lock rule of fast dispatch. When
introduced fast dispatch, there exists three methods
ms_fast_connect, ms_fast_accept,
supplement
On Tue, Dec 2, 2014 at 4:27 PM, Haomai Wang haomaiw...@gmail.com wrote:
Hi Gregoy and Sage,
I'm just writing Messenger's unit tests to ensure that SimpleMessenger
and AsyncMessenger behavior same and expected.
I think the most unclear thing is the lock rule of fast dispatch. When
1 - 100 of 225 matches
Mail list logo
