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
4082 zfs receive gets EFBIG from dmu_tx_hold_free()
Reviewed by: Eric Schrock <eric.schrock@delphix.com>
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Reviewed by: George Wilson <george.wilson@delphix.com>
Approved by: Richard Lowe <richlowe@richlowe.net>
References:
https://www.illumos.org/issues/4082illumos/illumos-gate@5253393b09
Ported-by: Richard Yao <ryao@gentoo.org>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #1775
4047 panic from dbuf_free_range() from dmu_free_object() while
doing zfs receive
Reviewed by: Adam Leventhal <ahl@delphix.com>
Reviewed by: George Wilson <george.wilson@delphix.com>
Approved by: Dan McDonald <danmcd@nexenta.com>
References:
https://www.illumos.org/issues/4047illumos/illumos-gate@713d6c2088
Ported-by: Richard Yao <ryao@gentoo.org>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #1775
Porting notes:
1. The exported symbol dmu_free_object() was renamed to
dmu_free_long_object() in Illumos.
3236 zio nop-write
Reviewed by: Matt Ahrens <matthew.ahrens@delphix.com>
Reviewed by: Adam Leventhal <ahl@delphix.com>
Reviewed by: Christopher Siden <chris.siden@delphix.com>
Approved by: Garrett D'Amore <garrett@damore.org>
References:
illumos/illumos-gate@80901aea8ehttps://www.illumos.org/issues/3236
Porting Notes
1. This patch is being merged dispite an increased instance of
https://www.illumos.org/issues/3113 being triggered by ztest.
Ported-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #1489
3741 zfs needs better comments
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Reviewed by: Eric Schrock <eric.schrock@delphix.com>
Approved by: Christopher Siden <christopher.siden@delphix.com>
References:
https://www.illumos.org/issues/3741illumos/illumos-gate@3e30c24aee
Ported-by: Richard Yao <ryao@gentoo.org>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #1775
3598 want to dtrace when errors are generated in zfs
Reviewed by: Dan Kimmel <dan.kimmel@delphix.com>
Reviewed by: Adam Leventhal <ahl@delphix.com>
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Approved by: Garrett D'Amore <garrett@damore.org>
References:
https://www.illumos.org/issues/3598illumos/illumos-gate@be6fd75a69
Ported-by: Richard Yao <ryao@gentoo.org>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #1775
Porting notes:
1. include/sys/zfs_context.h has been modified to render some new
macros inert until dtrace is available on Linux.
2. Linux-specific changes have been adapted to use SET_ERROR().
3. I'm NOT happy about this change. It does nothing but ugly
up the code under Linux. Unfortunately we need to take it to
avoid more merge conflicts in the future. -Brian
3603 panic from bpobj_enqueue_subobj()
3604 zdb should print bpobjs more verbosely
3871 GCC 4.5.3 does not like issue 3604 patch
Reviewed by: Henrik Mattson <henrik.mattson@delphix.com>
Reviewed by: Adam Leventhal <ahl@delphix.com>
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Garrett D'Amore <garrett@damore.org>
Approved by: Dan McDonald <danmcd@nexenta.com>
References:
https://www.illumos.org/issues/3603https://www.illumos.org/issues/3604https://www.illumos.org/issues/3871illumos/illumos-gate@d04756377d
Ported-by: Richard Yao <ryao@gentoo.org>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #1775
Note that the patch from Illumos issue 3871 is not accepted into Illumos
at the time of this writing. It is something that I wrote when porting
this. Documentation is in the Illumos issue.
3522 zfs module should not allow uninitialized variables
Reviewed by: Sebastien Roy <seb@delphix.com>
Reviewed by: Adam Leventhal <ahl@delphix.com>
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Approved by: Garrett D'Amore <garrett@damore.org>
References:
https://www.illumos.org/issues/3522illumos/illumos-gate@d5285cae91
Ported-by: Richard Yao <ryao@gentoo.org>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Porting notes:
1. ZFSOnLinux had already addressed many of these issues because of
its use of -Wall. However, the manner in which they were addressed
differed. The illumos fixes replace the ones previously made in
ZFSOnLinux to reduce code differences.
2. Part of the upstream patch made a small change to arc.c that might
address zfsonlinux/zfs#1334.
3. The initialization of aclsize in zfs_log_create() differs because
vsecp is a NULL pointer on ZFSOnLinux.
4. The changes to zfs_register_callbacks() were dropped because it
has diverged and needs to be resynced.
Currently there is no mechanism to inspect which dbufs are being
cached by the system. There are some coarse counters in arcstats
by they only give a rough idea of what's being cached. This patch
aims to improve the current situation by adding a new dbufs kstat.
When read this new kstat will walk all cached dbufs linked in to
the dbuf_hash. For each dbuf it will dump detailed information
about the buffer. It will also dump additional information about
the referenced arc buffer and its related dnode. This provides a
more complete view in to exactly what is being cached.
With this generic infrastructure in place utilities can be written
to post-process the data to understand exactly how the caching is
working. For example, the data could be processed to show a list
of all cached dnodes and how much space they're consuming. Or a
similar list could be generated based on dnode type. Many other
ways to interpret the data exist based on what kinds of questions
you're trying to answer.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Prakash Surya <surya1@llnl.gov>
3137 L2ARC compression
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Approved by: Dan McDonald <danmcd@nexenta.com>
References:
illumos/illumos-gate@aad02571bchttps://www.illumos.org/issues/3137http://wiki.illumos.org/display/illumos/L2ARC+Compression
Notes for Linux port:
A l2arc_nocompress module option was added to prevent the
compression of l2arc buffers regardless of how a dataset's
compression property is set. This allows the legacy behavior
to be preserved.
Ported by: James H <james@kagisoft.co.uk>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#1379
The dmu_prefetch, dmu_free_long_range, dmu_free_object,
dmu_prealloc, dmu_write_policy, and dmu_sync symbols have
been exported so they may be used by other modules.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
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>
When using a zvol to back a btrfs filesystem the btrfs mount
would hang. This was due to the bio completion callback used
in btrfs assuming that lower level drivers would never modify
the bio->bi_io_vecs after they were submitted via bio_submit().
If they are modified btrfs will miscalculate which pages need
to be unlocked resulting in a hang.
It's worth mentioning that other file systems such as ext[234]
and xfs work fine because they do not make the same assumption
in the bio completion callback.
The most straight forward way to fix the issue is to present
the semantics expected by btrfs. This is done by cloning the
bios attached to each request and then using the clones bvecs
to perform the required accounting. The clones are freed after
each read/write and the original unmodified bios are linked back
in to the request.
Signed-off-by: Chris Wedgwood <cw@f00f.org>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#469
Currently, ZIL blocks are spread over vdevs using hint block pointers
managed by the ZIL commit code and passed to metaslab_alloc(). Spreading
log blocks accross vdevs is important for performance: indeed, using
mutliple disks in parallel decreases the ZIL commit latency, which is
the main performance metric for synchronous writes. However, the current
implementation suffers from the following issues:
1) It would be best if the ZIL module was not aware of such low-level
details. They should be handled by the ZIO and metaslab modules;
2) Because the hint block pointer is managed per log, simultaneous
commits from multiple logs might use the same vdevs at the same time,
which is inefficient;
3) Because dmu_write() does not honor the block pointer hint, indirect
writes are not spread.
The naive solution of rotating the metaslab rotor each time a block is
allocated for the ZIL or dmu_sync() doesn't work in practice because the
first ZIL block to be written is actually allocated during the previous
commit. Consequently, when metaslab_alloc() decides the vdev for this
block, it will do so while a bunch of other allocations are happening at
the same time (from dmu_sync() and other ZILs). This means the vdev for
this block is chosen more or less at random. When the next commit
happens, there is a high chance (especially when the number of blocks
per commit is slightly less than the number of the disks) that one disk
will have to write two blocks (with a potential seek) while other disks
are sitting idle, which defeats spreading and increases the commit
latency.
This commit introduces a new concept in the metaslab allocator:
fastwrites. Basically, each top-level vdev maintains a counter
indicating the number of synchronous writes (from dmu_sync() and the
ZIL) which have been allocated but not yet completed. When the metaslab
is called with the FASTWRITE flag, it will choose the vdev with the
least amount of pending synchronous writes. If there are multiple vdevs
with the same value, the first matching vdev (starting from the rotor)
is used. Once metaslab_alloc() has decided which vdev the block is
allocated to, it updates the fastwrite counter for this vdev.
The rationale goes like this: when an allocation is done with
FASTWRITE, it "reserves" the vdev until the data is written. Until then,
all future allocations will naturally avoid this vdev, even after a full
rotation of the rotor. As a result, pending synchronous writes at a
given point in time will be nicely spread over all vdevs. This contrasts
with the previous algorithm, which is based on the implicit assumption
that blocks are written instantaneously after they're allocated.
metaslab_fastwrite_mark() and metaslab_fastwrite_unmark() are used to
manually increase or decrease fastwrite counters, respectively. They
should be used with caution, as there is no per-BP tracking of fastwrite
information, so leaks and "double-unmarks" are possible. There is,
however, an assert in the vdev teardown code which will fire if the
fastwrite counters are not zero when the pool is exported or the vdev
removed. Note that as stated above, marking is also done implictly by
metaslab_alloc().
ZIO also got a new FASTWRITE flag; when it is used, ZIO will pass it to
the metaslab when allocating (assuming ZIO does the allocation, which is
only true in the case of dmu_sync). This flag will also trigger an
unmark when zio_done() fires.
A side-effect of the new algorithm is that when a ZIL stops being used,
its last block can stay in the pending state (allocated but not yet
written) for a long time, polluting the fastwrite counters. To avoid
that, I've implemented a somewhat crude but working solution which
unmarks these pending blocks in zil_sync(), thus guaranteeing that
linguering fastwrites will get pruned at each sync event.
The best performance improvements are observed with pools using a large
number of top-level vdevs and heavy synchronous write workflows
(especially indirect writes and concurrent writes from multiple ZILs).
Real-life testing shows a 200% to 300% performance increase with
indirect writes and various commit sizes.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #1013
Differences between how paging is done on Solaris and Linux can cause
deadlocks if KM_SLEEP is used in any the following contexts.
* The txg_sync thread
* The zvol write/discard threads
* The zpl_putpage() VFS callback
This is because KM_SLEEP will allow for direct reclaim which may result
in the VM calling back in to the filesystem or block layer to write out
pages. If a lock is held over this operation the potential exists to
deadlock the system. To ensure forward progress all memory allocations
in these contexts must us KM_PUSHPAGE which disables performing any I/O
to accomplish the memory allocation.
Previously, this behavior was acheived by setting PF_MEMALLOC on the
thread. However, that resulted in unexpected side effects such as the
exhaustion of pages in ZONE_DMA. This approach touchs more of the zfs
code, but it is more consistent with the right way to handle these cases
under Linux.
This is patch lays the ground work for being able to safely revert the
following commits which used PF_MEMALLOC:
21ade34 Disable direct reclaim for z_wr_* threads
cfc9a5c Fix zpl_writepage() deadlock
eec8164 Fix ASSERTION(!dsl_pool_sync_context(tx->tx_pool))
Signed-off-by: Richard Yao <ryao@cs.stonybrook.edu>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #726
Expose the zfs_mdcomp_disable variable as a module option. This
can be used to disable compression of zfs meta data which is
enabled by default. This shouldn't need to be tuned but for
most workloads, however there may be very specific instances
where it makes sense to trade disk capacity for extra cpu cycles.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Keep counters for the various reasons that a thread may end up
in txg_wait_open() waiting on a new txg. This can be useful
when attempting to determine why a particular workload is
under performing.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Export additional symbols to make use of the DMU's zero-copy
API. This allows external modules to move data in to and out of
the ARC without incurring the cost of a memory copy.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Exported the required symbols to make use of the DMU's zero-copy
API. This allows external modules to move data in to and out of
the ARC without incurring the cost of a memory copy.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
The following warning was observed under normal operation. It's
not fatal but it's something to be addressed long term. Flag the
offending allocation with KM_NODEBUG to suppress the warning and
flag the call site.
SPL: Showing stack for process 21761
Pid: 21761, comm: iozone Tainted: P ----------------
2.6.32-71.14.1.el6.x86_64 #1
Call Trace:
[<ffffffffa05465a7>] spl_debug_dumpstack+0x27/0x40 [spl]
[<ffffffffa054a84d>] kmem_alloc_debug+0x11d/0x130 [spl]
[<ffffffffa05de166>] dmu_buf_hold_array_by_dnode+0xa6/0x4e0 [zfs]
[<ffffffffa05de825>] dmu_buf_hold_array+0x65/0x90 [zfs]
[<ffffffffa05de891>] dmu_read_uio+0x41/0xd0 [zfs]
[<ffffffffa0654827>] zfs_read+0x147/0x470 [zfs]
[<ffffffffa06644a2>] zpl_read_common+0x52/0x70 [zfs]
[<ffffffffa0664503>] zpl_read+0x43/0x70 [zfs]
[<ffffffff8116d905>] vfs_read+0xb5/0x1a0
[<ffffffff8116da41>] sys_read+0x51/0x90
[<ffffffff81013172>] system_call_fastpath+0x16/0x1b
For the moment we do not use dmu_write_pages() to write pages
directly in to a dmu object. It may be required at some point
in the future, but for now is simplest and cleanest to drop it.
It can be easily readded if/when needed.
These functions were dropped originally because I felt they would
need to be rewritten anyway to avoid using uios. However, this
patch readds then with they dea they can just be reworked and
the uio bits dropped.
Move xiou stat structures from a header to the dmu.c source as is
done with all the other kstat interfaces. This information is local
to dmu.c registered the xuio kstat and should stay that way.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Fix non-c90 compliant code, for the most part these changes
simply deal with where a particular variable is declared.
Under c90 it must alway be done at the very start of a block.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
