4045 zfs write throttle & i/o scheduler performance work
1. The ZFS i/o scheduler (vdev_queue.c) now divides i/os into 5 classes: sync
read, sync write, async read, async write, and scrub/resilver. The scheduler
issues a number of concurrent i/os from each class to the device. Once a class
has been selected, an i/o is selected from this class using either an elevator
algorithem (async, scrub classes) or FIFO (sync classes). The number of
concurrent async write i/os is tuned dynamically based on i/o load, to achieve
good sync i/o latency when there is not a high load of writes, and good write
throughput when there is. See the block comment in vdev_queue.c (reproduced
below) for more details.
2. The write throttle (dsl_pool_tempreserve_space() and
txg_constrain_throughput()) is rewritten to produce much more consistent delays
when under constant load. The new write throttle is based on the amount of
dirty data, rather than guesses about future performance of the system. When
there is a lot of dirty data, each transaction (e.g. write() syscall) will be
delayed by the same small amount. This eliminates the "brick wall of wait"
that the old write throttle could hit, causing all transactions to wait several
seconds until the next txg opens. One of the keys to the new write throttle is
decrementing the amount of dirty data as i/o completes, rather than at the end
of spa_sync(). Note that the write throttle is only applied once the i/o
scheduler is issuing the maximum number of outstanding async writes. See the
block comments in dsl_pool.c and above dmu_tx_delay() (reproduced below) for
more details.
This diff has several other effects, including:
* the commonly-tuned global variable zfs_vdev_max_pending has been removed;
use per-class zfs_vdev_*_max_active values or zfs_vdev_max_active instead.
* the size of each txg (meaning the amount of dirty data written, and thus the
time it takes to write out) is now controlled differently. There is no longer
an explicit time goal; the primary determinant is amount of dirty data.
Systems that are under light or medium load will now often see that a txg is
always syncing, but the impact to performance (e.g. read latency) is minimal.
Tune zfs_dirty_data_max and zfs_dirty_data_sync to control this.
* zio_taskq_batch_pct = 75 -- Only use 75% of all CPUs for compression,
checksum, etc. This improves latency by not allowing these CPU-intensive tasks
to consume all CPU (on machines with at least 4 CPU's; the percentage is
rounded up).
--matt
APPENDIX: problems with the current i/o scheduler
The current ZFS i/o scheduler (vdev_queue.c) is deadline based. The problem
with this is that if there are always i/os pending, then certain classes of
i/os can see very long delays.
For example, if there are always synchronous reads outstanding, then no async
writes will be serviced until they become "past due". One symptom of this
situation is that each pass of the txg sync takes at least several seconds
(typically 3 seconds).
If many i/os become "past due" (their deadline is in the past), then we must
service all of these overdue i/os before any new i/os. This happens when we
enqueue a batch of async writes for the txg sync, with deadlines 2.5 seconds in
the future. If we can't complete all the i/os in 2.5 seconds (e.g. because
there were always reads pending), then these i/os will become past due. Now we
must service all the "async" writes (which could be hundreds of megabytes)
before we service any reads, introducing considerable latency to synchronous
i/os (reads or ZIL writes).
Notes on porting to ZFS on Linux:
- zio_t gained new members io_physdone and io_phys_children. Because
object caches in the Linux port call the constructor only once at
allocation time, objects may contain residual data when retrieved
from the cache. Therefore zio_create() was updated to zero out the two
new fields.
- vdev_mirror_pending() relied on the depth of the per-vdev pending queue
(vq->vq_pending_tree) to select the least-busy leaf vdev to read from.
This tree has been replaced by vq->vq_active_tree which is now used
for the same purpose.
- vdev_queue_init() used the value of zfs_vdev_max_pending to determine
the number of vdev I/O buffers to pre-allocate. That global no longer
exists, so we instead use the sum of the *_max_active values for each of
the five I/O classes described above.
- The Illumos implementation of dmu_tx_delay() delays a transaction by
sleeping in condition variable embedded in the thread
(curthread->t_delay_cv). We do not have an equivalent CV to use in
Linux, so this change replaced the delay logic with a wrapper called
zfs_sleep_until(). This wrapper could be adopted upstream and in other
downstream ports to abstract away operating system-specific delay logic.
- These tunables are added as module parameters, and descriptions added
to the zfs-module-parameters.5 man page.
spa_asize_inflation
zfs_deadman_synctime_ms
zfs_vdev_max_active
zfs_vdev_async_write_active_min_dirty_percent
zfs_vdev_async_write_active_max_dirty_percent
zfs_vdev_async_read_max_active
zfs_vdev_async_read_min_active
zfs_vdev_async_write_max_active
zfs_vdev_async_write_min_active
zfs_vdev_scrub_max_active
zfs_vdev_scrub_min_active
zfs_vdev_sync_read_max_active
zfs_vdev_sync_read_min_active
zfs_vdev_sync_write_max_active
zfs_vdev_sync_write_min_active
zfs_dirty_data_max_percent
zfs_delay_min_dirty_percent
zfs_dirty_data_max_max_percent
zfs_dirty_data_max
zfs_dirty_data_max_max
zfs_dirty_data_sync
zfs_delay_scale
The latter four have type unsigned long, whereas they are uint64_t in
Illumos. This accommodates Linux's module_param() supported types, but
means they may overflow on 32-bit architectures.
The values zfs_dirty_data_max and zfs_dirty_data_max_max are the most
likely to overflow on 32-bit systems, since they express physical RAM
sizes in bytes. In fact, Illumos initializes zfs_dirty_data_max_max to
2^32 which does overflow. To resolve that, this port instead initializes
it in arc_init() to 25% of physical RAM, and adds the tunable
zfs_dirty_data_max_max_percent to override that percentage. While this
solution doesn't completely avoid the overflow issue, it should be a
reasonable default for most systems, and the minority of affected
systems can work around the issue by overriding the defaults.
- Fixed reversed logic in comment above zfs_delay_scale declaration.
- Clarified comments in vdev_queue.c regarding when per-queue minimums take
effect.
- Replaced dmu_tx_write_limit in the dmu_tx kstat file
with dmu_tx_dirty_delay and dmu_tx_dirty_over_max. The first counts
how many times a transaction has been delayed because the pool dirty
data has exceeded zfs_delay_min_dirty_percent. The latter counts how
many times the pool dirty data has exceeded zfs_dirty_data_max (which
we expect to never happen).
- The original patch would have regressed the bug fixed in
zfsonlinux/zfs@c418410, which prevented users from setting the
zfs_vdev_aggregation_limit tuning larger than SPA_MAXBLOCKSIZE.
A similar fix is added to vdev_queue_aggregate().
- In vdev_queue_io_to_issue(), dynamically allocate 'zio_t search' on the
heap instead of the stack. In Linux we can't afford such large
structures on the stack.
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Adam Leventhal <ahl@delphix.com>
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Reviewed by: Ned Bass <bass6@llnl.gov>
Reviewed by: Brendan Gregg <brendan.gregg@joyent.com>
Approved by: Robert Mustacchi <rm@joyent.com>
References:
http://www.illumos.org/issues/4045illumos/illumos-gate@69962b5647
Ported-by: Ned Bass <bass6@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#1913
3742 zfs comments need cleaner, more consistent style
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Eric Schrock <eric.schrock@delphix.com>
Approved by: Christopher Siden <christopher.siden@delphix.com>
References:
https://www.illumos.org/issues/3742illumos/illumos-gate@f717074149
Ported-by: Richard Yao <ryao@gentoo.org>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #1775
Porting notes:
1. The change to zfs_vfsops.c was dropped because it involves
zfs_mount_label_policy, which does not exist in the Linux port.
This change adds a new kstat to gain some visibility into the amount of
time spent in each call to dmu_tx_assign. A histogram is exported via
a new dmu_tx_assign_histogram-$POOLNAME file. The information contained
in this histogram is the frequency dmu_tx_assign took to complete given
an interval range. For example, given the below histogram file:
$ cat /proc/spl/kstat/zfs/dmu_tx_assign_histogram-tank
12 1 0x01 32 1536 19792068076691 20516481514522
name type data
1 us 4 859
2 us 4 252
4 us 4 171
8 us 4 2
16 us 4 0
32 us 4 2
64 us 4 0
128 us 4 0
256 us 4 0
512 us 4 0
1024 us 4 0
2048 us 4 0
4096 us 4 0
8192 us 4 0
16384 us 4 0
32768 us 4 1
65536 us 4 1
131072 us 4 1
262144 us 4 4
524288 us 4 0
1048576 us 4 0
2097152 us 4 0
4194304 us 4 0
8388608 us 4 0
16777216 us 4 0
33554432 us 4 0
67108864 us 4 0
134217728 us 4 0
268435456 us 4 0
536870912 us 4 0
1073741824 us 4 0
2147483648 us 4 0
one can see most calls to dmu_tx_assign completed in 32us or less, but a
few outliers did not. Specifically, 4 of the calls took between 262144us
and 131072us. This information is difficult, if not impossible, to gather
without this change.
Signed-off-by: Prakash Surya <surya1@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#1584
Reviewed by: Matt Ahrens <matthew.ahrens@delphix.com>
Reviewed by: Eric Schrock <eric.schrock@delphix.com>
Reviewed by: Christopher Siden <chris.siden@delphix.com>
Approved by: Garrett D'Amore <garrett@damore.org>
NOTES: This patch has been reworked from the original in the
following ways to accomidate Linux ZFS implementation
*) Usage of the cyclic interface was replaced by the delayed taskq
interface. This avoids the need to implement new compatibility
code and allows us to rely on the existing taskq implementation.
*) An extern for zfs_txg_synctime_ms was added to sys/dsl_pool.h
because declaring externs in source files as was done in the
original patch is just plain wrong.
*) Instead of panicing the system when the deadman triggers a
zevent describing the blocked vdev and the first pending I/O
is posted. If the panic behavior is desired Linux provides
other generic methods to panic the system when threads are
observed to hang.
*) For reference, to delay zios by 30 seconds for testing you can
use zinject as follows: 'zinject -d <vdev> -D30 <pool>'
References:
illumos/illumos-gate@283b84606bhttps://www.illumos.org/issues/3246
Ported-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#1396
2619 asynchronous destruction of ZFS file systems
2747 SPA versioning with zfs feature flags
Reviewed by: Matt Ahrens <mahrens@delphix.com>
Reviewed by: George Wilson <gwilson@delphix.com>
Reviewed by: Richard Lowe <richlowe@richlowe.net>
Reviewed by: Dan Kruchinin <dan.kruchinin@gmail.com>
Approved by: Eric Schrock <Eric.Schrock@delphix.com>
References:
illumos/illumos-gate@53089ab7c8illumos/illumos-gate@ad135b5d64
illumos changeset: 13700:2889e2596bd6
https://www.illumos.org/issues/2619https://www.illumos.org/issues/2747
NOTE: The grub specific changes were not ported. This change
must be made to the Linux grub packages.
Ported-by: Brian Behlendorf <behlendorf1@llnl.gov>
Create a kstat file which contains useful statistics about the
last N txgs processed. This can be helpful when analyzing pool
performance. The new KSTAT_TYPE_TXG type was added for this
purpose and it tracks the following statistics per-txg.
txg - Unique txg number
state - State (O)pen/(Q)uiescing/(S)yncing/(C)ommitted
birth; - Creation time
nread - Bytes read
nwritten; - Bytes written
reads - IOPs read
writes - IOPs write
open_time; - Length in nanoseconds the txg was open
quiesce_time - Length in nanoseconds the txg was quiescing
sync_time; - Length in nanoseconds the txg was syncing
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
I appologize in advance why to many things ended up in this commit.
When it could be seperated in to a whole series of commits teasing
that all apart now would take considerable time and I'm not sure
there's much merrit in it. As such I'll just summerize the intent
of the changes which are all (or partly) in this commit. Broadly
the intent is to remove as much Solaris specific code as possible
and replace it with native Linux equivilants. More specifically:
1) Replace all instances of zfsvfs_t with zfs_sb_t. While the
type is largely the same calling it private super block data
rather than a zfsvfs is more consistent with how Linux names
this. While non critical it makes the code easier to read when
your thinking in Linux friendly VFS terms.
2) Replace vnode_t with struct inode. The Linux VFS doesn't have
the notion of a vnode and there's absolutely no good reason to
create one. There are in fact several good reasons to remove it.
It just adds overhead on Linux if we were to manage one, it
conplicates the code, and it likely will lead to bugs so there's
a good change it will be out of date. The code has been updated
to remove all need for this type.
3) Replace all vtype_t's with umode types. Along with this shift
all uses of types to mode bits. The Solaris code would pass a
vtype which is redundant with the Linux mode. Just update all the
code to use the Linux mode macros and remove this redundancy.
4) Remove using of vn_* helpers and replace where needed with
inode helpers. The big example here is creating iput_aync to
replace vn_rele_async. Other vn helpers will be addressed as
needed but they should be be emulated. They are a Solaris VFS'ism
and should simply be replaced with Linux equivilants.
5) Update znode alloc/free code. Under Linux it's common to
embed the inode specific data with the inode itself. This removes
the need for an extra memory allocation. In zfs this information
is called a znode and it now embeds the inode with it. Allocators
have been updated accordingly.
6) Minimal integration with the vfs flags for setting up the
super block and handling mount options has been added this
code will need to be refined but functionally it's all there.
This will be the first and last of these to large to review commits.
One of the neat tricks an autoconf style project is capable of
is allow configurion/building in a directory other than the
source directory. The major advantage to this is that you can
build the project various different ways while making changes
in a single source tree.
For example, this project is designed to work on various different
Linux distributions each of which work slightly differently. This
means that changes need to verified on each of those supported
distributions perferably before the change is committed to the
public git repo.
Using nfs and custom build directories makes this much easier.
I now have a single source tree in nfs mounted on several different
systems each running a supported distribution. When I make a
change to the source base I suspect may break things I can
concurrently build from the same source on all the systems each
in their own subdirectory.
wget -c http://github.com/downloads/behlendorf/zfs/zfs-x.y.z.tar.gz
tar -xzf zfs-x.y.z.tar.gz
cd zfs-x-y-z
------------------------- run concurrently ----------------------
<ubuntu system> <fedora system> <debian system> <rhel6 system>
mkdir ubuntu mkdir fedora mkdir debian mkdir rhel6
cd ubuntu cd fedora cd debian cd rhel6
../configure ../configure ../configure ../configure
make make make make
make check make check make check make check
This change also moves many of the include headers from individual
incude/sys directories under the modules directory in to a single
top level include directory. This has the advantage of making
the build rules cleaner and logically it makes a bit more sense.
