Can I know how to configure a SSD to be used for L2arc ? Basically I want to
improve read performance.
Read the documentation, specifically the section titled;
Creating a ZFS Storage PoolWith Cache Devices
To increase write performance, will SSD for Zil help ? As I read on forums,
Zil
So try unbinding the mac threads; it may help you here.
How do I do that? All I can find on interrupt fencing and the like is to
simply set certain processors to no-intr, which moves all of the
interrupts and it doesn't prevent the xcall storm choosing to affect
these CPUs either…
In
I can't help but be curious about something, which perhaps you verified but
did not post.
What the data here shows is;
- CPU 31 is buried in the kernel (100% sys).
- CPU 31 is handling a moderate-to-high rate of xcalls.
What the data does not prove empirically is that the 100% sys time of
CPU
THE PROBLEM - Linux is 15% sys, 55% usr,
Solaris is 30% sys, 70% usr, running the same workload,
doing the same amount of work. delivering the same level
of performance. Please validate that problem statement.
On Mar 25, 2012, at 9:51 PM, Aubrey Li wrote:
On Mon, Mar 26, 2012 at 4:18 AM, Jim
If you're chasing CPU utilization, specifically %sys (time in the kernel),
I would start with a time-based kernel profile.
#dtrace -n 'profile-997hz /arg0/ { @[stack()] = count(); } tick-60sec {
trunc(@, 20); printa(@0; }'
I would be curious to see where the CPU cycles are being consumed first,
Does this reveal anything;
dtrace -n 'syscall::*write:entry /fds[arg0].fi_fs == zfs/ {
@[execname,fds[arg0].fi_pathname]=count(); }'
On Jun 11, 2011, at 9:32 AM, Jim Klimov wrote:
While looking over iostats from various programs, I see that
my OS HDD is busy writing, about 2Mb/sec stream
]-fi_pathname] = count(); }'
On Jun 11, 2011, at 12:34 PM, Jim Klimov wrote:
2011-06-11 19:16, Jim Mauro пишет:
Does this reveal anything;
dtrace -n 'syscall::*write:entry /fds[arg0].fi_fs == zfs/ {
@[execname,fds[arg0].fi_pathname]=count(); }'
Alas, not much.
# time dtrace -n 'syscall
in IOPS
and/or bytes being written?
On Jun 11, 2011, at 1:00 PM, Jim Klimov wrote:
2011-06-11 20:42, Jim Mauro пишет:
Well we may have missed something, because that dtrace will
only capture write(2) and pwrite(2) - whatever is generating the writes
may be using another interface (writev(2
This may be interesting also (still fumbling...);
dtrace -n 'fbt:zfs:zio_write:entry,
fbt:zfs:zio_rewrite:entry,fbt:zfs:zio_write_override:entry {
@[probefunc,stack()] = count(); }'
On Jun 11, 2011, at 1:00 PM, Jim Klimov wrote:
2011-06-11 20:42, Jim Mauro пишет:
Well we may have missed
With ZFS, Solaris 10 Update 9, is it possible to
detach configured log devices from a zpool?
I have a zpool with 3 F20 mirrors for the ZIL. They're
coming up corrupted. I want to detach them, remake
the devices and reattach them to the zpool.
Thanks
/jim
Hi Jim - cross-posting to zfs-discuss, because 20X is, to say the least,
compelling.
Obviously, it would be awesome if we had the opportunity to whittle-down which
of
the changes made this fly, or if it was a combination of the changes.
Looking at them individually
set
There is nothing in here that requires zfs confidential.
cross-posted to zfs discuss.
On Oct 21, 2010, at 3:37 PM, Jim Nissen wrote:
Cross-posting.
Original Message
Subject: Performance problems due to smaller ZFS recordsize
Date: Thu, 21 Oct 2010 14:00:42 -0500
ZFS does not use segmap.
The ZFS ARC (Adaptive Replacement Cache) will consume what's
available, memory-wise, based on the workload. There's an upper
limit if zfs_arc_max has not been set, but I forget what it is.
If other memory consumers (applications, other kernel subsystems)
need memory,
Using ZFS for Oracle can be configured to deliver very good performance.
Depending on what your priorities are in terms of critical metrics, keep
in mind
that the most performant solution is to use Oracle ASM on raw disk devices.
That is not intended to imply anything negative about ZFS or UFS.
Hi Anthony -
I don't get this. How does the presence (or absence) of the ARC change
the methodology for doing memory capacity planning?
Memory capacity planning is all about identifying and measuring consumers.
Memory consumers;
- The kernel.
- User processes.
- The ZFS ARC, which is
Posting to zfs-discuss. There's no reason this needs to be
kept confidential.
5-disk RAIDZ2 - doesn't that equate to only 3 data disks?
Seems pointless - they'd be much better off using mirrors,
which is a better choice for random IO...
Looking at this now...
/jim
Jeff Savit wrote:
Hi all,
I'm cross-posting to zfs-discuss, as this is now more of a ZFS
query than a dtrace query at this point, and I'm not sure if all the ZFS
experts are listening on dtrace-discuss (although they probably
are... :^).
The only thing that jumps out at me is the ARC size - 53.4GB, or
most of your 64GB
(posted to zfs-discuss)
Hmmm...this is nothing in terms of load.
So you say that the system becomes sluggish/unresponsive
periodically, and you noticed the xcall storm when that
happens, correct?
Refresh my memory - what is the frequency and duration
of the sluggish cycles?
Could you capture a
Cross-posting to zfs-discuss. This does not need to be on the
confidential alias. It's a performance query - there's nothing
confidential in here. Other folks post performance queries to
zfs-discuss
Forget %b - it's useless.
It's not the bandwidth that's hurting you, it's the IOPS.
One of
Bob - Have you filed a bug on this issue?
I am not up to speed on this thread, so I can
not comment on whether or not there is a bug
here, but you seem to have a test case and supporting
data. Filing a bug will get the attention of ZFS
engineering.
Thanks,
/jim
Bob Friesenhahn wrote:
On Mon,
Posting this back to zfs-discuss.
Roland's test case (below) is a single threaded sequential write
followed by a single threaded sequential read. His bandwidth
goes from horrible (~2MB/sec) to expected (~30MB/sec)
when prefetch is disabled. This is with relatively recent nv bits (nv110).
Cross posting to zfs-discuss.
By my math, here's what you're getting;
4.6MB/sec on writes to ZFS.
2.2MB/sec on reads from ZFS.
90MB/sec on read from block device.
What is c0t1d0 - I assume it's a hardware RAID LUN,
but how many disks, and what type of LUN?
What version of Solaris (cat
Cross-posting to the public ZFS discussion alias.
There's nothing here that requires confidentiallity, and
the public alias is a much broader audience with a larger
number of experienced ZFS users...
As to the issue - what is the free space disparity
across the pools? Is the one particular
This problem only manifests itself when dealing with many small files
over NFS. There is no throughput problem with the network.
But there could be a _latency_ issue with the network.
[snip]
I've done my homework on this issue, I've ruled out the network as an
issue, as well as the NFS
You have SSD's for the ZIL (logzilla) enabled, and ZIL IO
is what is hurting your performance...Hmmm
I'll ask the stupid question (just to get it out of the way) - is
it possible that the logzilla is undersized?
Did you gather data using Richard Elling's zilstat (included below)?
Thanks,
Sogranted, tank is about 77% full (not to split hairs ;^),
but in this case, 23% is 640GB of free space. I mean, it's
not like 15 years ago when a file system was 2GB total,
and 23% free meant a measely 460MB to allocate from.
640GB is a lot of space, and our largest writes are less
than 5MB.
Multiple Thors (more than 2?), with performance problems.
Maybe it's the common demnominator - the network.
Can you run local ZFS IO loads and determine if performance
is expected when NFS and the network are out of the picture?
Thanks,
/jim
Greg Mason wrote:
So, I'm still beating my head
on x64 - it will panic your system.
None of this has anything to do with ZFS, which uses a completely different
mechanism for caching (the ZFS ARC).
Thanks,
/jim
What is what I heard Jim Mauro tell us. I recall feeling a bit
disturbed when I heard it. If it is true, perhaps it applies only
So I'm really exposing my ignorance here, but...
You wrote /... if you wish to keep your snapshots.../...
I never mentioned snapshots, thus you
introduced the use of a ZFS snapshot as a method of doing what
I wish to do. And yes, snapshots and send are in the manual, and
I read about them.
I
http://www.solarisinternals.com/wiki/index.php/ZFS_Evil_Tuning_Guide#Disabling_the_ZIL_.28Don.27t.29
The above link shows how to disable to ZIL for testing purposes (it's
not generally recommended to
keep it disabled in production).
As to the putpack schedule of recent ZFS features into
Would you two please SHUT THE F$%K UP.
Dear God, my kids don't go own like this.
Please - let it die already.
Thanks very much.
/jim
can you guess? wrote:
Hello can,
Thursday, December 13, 2007, 12:02:56 AM, you wrote:
cyg On the other hand, there's always the
possibility that someone
If the question is can Oracle files (datafiles, log files, etc) exist
on a ZFS, the
answer is absolutely yes. More simply put, can you configure you Oracle
database
on ZFS - absolutely.
The question, as stated, is confusing, because the term compatible can
have pretty
broad meaning. So, I
Where does the win come from with directI/O? Is it 1), 2), or some
combination? If its a combination, what's the percentage of each
towards the win?
That will vary based on workload (I know, you already knew that ... :^).
Decomposing the performance win between what is gained as a
Hey Roch -
We do not retain 2 copies of the same data.
If the DB cache is made large enough to consume most of memory,
the ZFS copy will quickly be evicted to stage other I/Os on
their way to the DB cache.
What problem does that pose ?
Can't answer that question empirically, because we
Hi Neel - Thanks for pushing this out. I've been tripping over this for
a while.
You can instrument zfs_read() and zfs_write() to reliably track filenames:
#!/usr/sbin/dtrace -s
#pragma D option quiet
zfs_read:entry,
zfs_write:entry
{
printf(%s of %s\n,probefunc,
What sayeth the ZFS team regarding the use of a stable DTrace provider
with their file system?
For the record, the above has a tone to it that I really did not intend
(antagonistic?), so
I had a good chat with Roch about this. The file pathname is derived via
a translator
from the
files, and it seems to work
/jim
Neelakanth Nadgir wrote:
Jim I can't use zfs_read/write as the file is mmap()'d so no read/write!
-neel
On Sep 26, 2007, at 5:07 AM, Jim Mauro [EMAIL PROTECTED] wrote:
Hi Neel - Thanks for pushing this out. I've been tripping over this
for a while
Hey Max - Check out the on-disk specification document at
http://opensolaris.org/os/community/zfs/docs/.
Page 32 illustration shows the rootbp pointing to a dnode_phys_t
object (the first member of a objset_phys_t data structure).
The source code indicates ub_rootbp is a blkptr_t, which
About 2 years ago I was able to get a little closer to the patent
litigation process,
by way of giving a deposition in litigation that was filed against Sun
and Apple
(and has been settled).
Apparently, there's an entire sub-economy built on patent litigation
among the
technology players.
I'll take a look at this. ZFS provides outstanding sequential IO performance
(both read and write). In my testing, I can essentially sustain
hardware speeds
with ZFS on sequential loads. That is, assuming 30-60MB/sec per disk
sequential
IO capability (depending on hitting inner or out
Is the referenced Laminated Handout on slide 3 available anywhere in any
form electronically?
If not, I'd be happy to create an electronic copy and make it pubically
available.
Thanks,
/jim
Joy Marshall wrote:
It's taken a while but at last we have been able to post the ZFS Under the
Hello -
I think L4 still needs to evolve. BTW, i believe microkernels is the
_right_ way and L4 is a first step in that direction.
Perhaps you could elaborate on this? I thought the microkernel debate ended
in the 1990s, in terms of being a compelling technology direction for kernel
I'm not sure I understand the question.
Virtual machines are built by either running a virtualization technology
in a host operating systems, such as running VMware Workstation in
Linux, running Parallels in Mac OS X, Linux or Windows, etc.
These are sometimes referred to as Type II VMMs, where
] writes:
Jim Mauro wrote:
All righty...I set c_max to 512MB, c to 512MB, and p to 256MB...
arc::print -tad
{
...
c02e29e8 uint64_t size = 0t299008
c02e29f0 uint64_t p = 0t16588228608
c02e29f8 uint64_t c = 0t33176457216
(I'm probably not the best person to answer this, but that has never
stopped me
before, and I need to give Richard Elling a little more time to get the
Goats, Cows
and Horses fed, sip his morning coffee, and offer a proper response...)
Would it benefit us to have the disk be setup as a raidz
http://www.cnn.com/2007/US/03/20/lost.data.ap/index.html
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
no data in those thousand's of file
systems yet.
Richard Elling wrote:
Jim Mauro wrote:
(I'm probably not the best person to answer this, but that has never
stopped me
before, and I need to give Richard Elling a little more time to get
the Goats, Cows
and Horses fed, sip his morning coffee
c02e2a08 uint64_t c_max = 0t1070318720
. . .
Perhaps c_max does not do what I think it does?
Thanks,
/jim
Jim Mauro wrote:
Running an mmap-intensive workload on ZFS on a X4500, Solaris 10 11/06
(update 3). All file IO is mmap(file), read memory segment, unmap, close.
Tweaked the arc
the ARC size because for mmap-intensive workloads,
it seems to hurt more than help (although, based on experiments up to this
point, it's not hurting a lot).
I'll do another reboot, and run it all down for you serially...
/jim
Thanks,
-j
On Thu, Mar 15, 2007 at 06:57:12PM -0400, Jim Mauro wrote
Following a reboot:
arc::print -tad
{
. . .
c02e29e8 uint64_t size = 0t299008
c02e29f0 uint64_t p = 0t16588228608
c02e29f8 uint64_t c = 0t33176457216
c02e2a00 uint64_t c_min = 0t1070318720
c02e2a08 uint64_t c_max = 0t33176457216
. . .
}
another reboot, and run it all down for you serially...
/jim
Thanks,
-j
On Thu, Mar 15, 2007 at 06:57:12PM -0400, Jim Mauro wrote:
ARC_mru::print -d size lsize
size = 0t10224433152
lsize = 0t10218960896
ARC_mfu::print -d size lsize
, 2007 at 06:57:12PM -0400, Jim Mauro wrote:
ARC_mru::print -d size lsize
size = 0t10224433152
lsize = 0t10218960896
ARC_mfu::print -d size lsize
size = 0t303450112
lsize = 0t289998848
ARC_anon::print -d size
You don't honestly, really, reasonably, expect someone, anyone, to look
at the stack
trace of a few hundred threads, and post something along the lines of
This is what is
wrong with your NFS server.Do you? Without any other information at
all?
We're here to help, but please reset your
Could be NFS synchronous semantics on file create (followed by
repeated flushing of the write cache). What kind of storage are you
using (feel free to send privately if you need to) - is it a thumper?
It's not clear why NFS-enforced synchronous semantics would induce
different behavior than
Hey Ben - I need more time to look at this and connect some dots,
but real quick
Some nfsstat data that we could use to potentially correlate to the local
server activity would be interesting. zfs_create() seems to be the
heavy hitter, but a periodic kernel profile (especially if we can
I need to read through this more thoroughly to get my head around it, but
on my first pass, what jumps out at me is that something significant
_changed_ in terms of application behavior with the introduction of ZFS.
I'm saying that that is a bad thing, or a good thing, but it is an
important
I agree with Greg - For ZFS, I'd recommend a larger number of raidz
luns, with a smaller number
of disks per LUN, up to 6 disks per raidz lun.
This will more closely align with performance best practices, so it
would be cool to find
common ground in terms of a sweet-spot for performance and
http://www.tech-recipes.com/solaris_system_administration_tips1446.html
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
58 matches
Mail list logo