Previously, ZFS scaled maxinflight_bytes based on total number of
disks in the pool. A 3-wide mirror was receiving a queue depth of 3
disks, which it should not, since it reads from all the disks inside.
For wide raidz the situation was slightly better, but still a 3-wide
raidz1 received a depth of 3 disks instead of 2.
The new code counts only unique data disks, i.e. 1 disk for mirrors
and non-parity disks for raidz/draid. For draid the math is still
imperfect, since vdev_get_nparity() returns number of parity disks
per group, not per vdev, but still some better than it was.
This should slightly reduce scrub influence on payload for some pool
topologies by avoiding excessive queuing.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Signed-off-by: Alexander Motin <mav@FreeBSD.org>
Sponsored-By: iXsystems, Inc.
Closing #12046
While use of dynamic taskqs allows to reduce number of idle threads,
hardcoded 8 taskqs of each kind is a big overkill for small systems,
complicating CPU scheduling, increasing I/O reorder, etc, while
providing no real locking benefits, just not needed there.
On another side, 12*8 worker threads per kind are able to overload
almost any system nowadays. For example, pool of several fast SSDs
with SHA256 checksum makes system barely responsive during scrub, or
with dedup enabled barely responsive during large file deletion.
To address both problems this patch introduces ZTI_SCALE macro, alike
to ZTI_BATCH, but with multiple taskqs, depending on number of CPUs,
to be used in places where lock scalability is needed, while request
ordering is not so much. The code is made to create new taskq for
~6 worker threads (less for small systems, but more for very large)
up to 80% of CPU cores (previous 75% was not good for rounding down).
Both number of threads and threads per taskq are now tunable in case
somebody really wants to use all of system power for ZFS.
While obviously some benchmarks show small peak performance reduction
(not so big really, especially on systems with SMT, where use of the
second threads does not give as much performance as the first ones),
they also show dramatic latency reduction and much more smooth user-
space operation in case of high CPU usage by ZFS.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Alexander Motin <mav@FreeBSD.org>
Sponsored-By: iXsystems, Inc.
Closes#11966
zfs_zevent_console committed multiple printk()s per line without
properly continuing them ‒ a single event could easily be fragmented
across over thirty lines, making it useless for direct application
zfs_zevent_cols exists purely to wrap the output from zfs_zevent_console
The niche this was supposed to fill can be better served by something
akin to the all-syslog ZEDLET
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ahelenia Ziemiańska <nabijaczleweli@nabijaczleweli.xyz>
Closes#7082Closes#11996
Changed the default specified for zfs_dbgmsg_enable, added
clarification of interaction with zfs_flags.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Ryan Moeller <ryan@ixsystems.com>
Signed-off-by: Rich Ercolani <rincebrain@gmail.com>
Closes#11984Closes#11986
Traversal code, traverse_visitbp() does visit blocks recursively.
Indirect (Non L0) Block of size 128k could contain, 1024 block pointers
of 128 bytes. In case of full traverse OR incremental traverse, where
all blocks were modified, it could traverse large number of blocks
pointed by indirect. Traversal code does issue prefetch of blocks
traversed below indirect. This could result into large number of
async reads queued on vdev queue. So, account for prefetch issued for
blocks pointed by indirect and limit max prefetch in one go.
Module Param:
zfs_traverse_indirect_prefetch_limit: Limit of prefetch while traversing
an indirect block.
Local counters:
prefetched: Local counter to account for number prefetch done.
pidx: Index for which next prefetch to be issued.
ptidx: Index at which next prefetch to be triggered.
Keep "ptidx" somewhere in the middle of blocks prefetched, so that
blocks prefetch read gets the enough time window before their demand
read is issued.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Signed-off-by: Jitendra Patidar <jitendra.patidar@nutanix.com>
Closes#11802Closes#11803
It's been observed in the CI that the required 25% of obsolete bytes
in the mapping can be to high a threshold for this test resulting in
condensing never being triggered and a test failure. To prevent these
failures make the existing zfs_condense_indirect_obsolete_pct tuning
available so the obsolete percentage can be reduced from 25% to 5%
during this test.
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Reviewed-by: George Melikov <mail@gmelikov.ru>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#11869
Just as delay zevents can flood the zevent pipe when a vdev becomes
unresponsive, so do the deadman zevents.
Ratelimit deadman zevents according to the same tunable as for delay
zevents.
Enable deadman tests on FreeBSD and add a test for deadman event
ratelimiting.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Don Brady <don.brady@delphix.com>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes#11786
Correct an assortment of typos throughout the code base.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Signed-off-by: Andrea Gelmini <andrea.gelmini@gelma.net>
Closes#11774
The lower bound for this scaling to too low and the upper bound is too
high. Use a fixed default length of 512 instead, which is a reasonable
value on any system.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes#11822
To make use of zfs_refcount_held tunable it should be a module
parameter in open-zfs. Also, since the macros will auto-generate OS
specific tunables, removed the existing zfs_refcount_held reference
in module/os/freebsd/zfs/sysctl_os.c.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Reviewed-by: Allan Jude <allan@klarasystems.com>
Signed-off-by: Don Brady <don.brady@delphix.com>
Closes#11753
On Linux increase the maximum allowed size of the src nvlist which
can be passed to the /dev/zfs ioctl. Originally, this was set
to a maximum of KMALLOC_MAX_SIZE (4M) because it was kmalloc'd.
Since that time it's been converted to a vmalloc so that's no
longer a hard limit, and it's desirable for `zfs send/recv` to
allow larger nvlists so more snapshots can be sent at once.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#6572Closes#11638
Mixing ZIL and normal allocations has several problems:
1. The ZIL allocations are allocated, written to disk, and then a few
seconds later freed. This leaves behind holes (free segments) where the
ZIL blocks used to be, which increases fragmentation, which negatively
impacts performance.
2. When under moderate load, ZIL allocations are of 128KB. If the pool
is fairly fragmented, there may not be many free chunks of that size.
This causes ZFS to load more metaslabs to locate free segments of 128KB
or more. The loading happens synchronously (from zil_commit()), and can
take around a second even if the metaslab's spacemap is cached in the
ARC. All concurrent synchronous operations on this filesystem must wait
while the metaslab is loading. This can cause a significant performance
impact.
3. If the pool is very fragmented, there may be zero free chunks of
128KB or more. In this case, the ZIL falls back to txg_wait_synced(),
which has an enormous performance impact.
These problems can be eliminated by using a dedicated log device
("slog"), even one with the same performance characteristics as the
normal devices.
This change sets aside one metaslab from each top-level vdev that is
preferentially used for ZIL allocations (vdev_log_mg,
spa_embedded_log_class). From an allocation perspective, this is
similar to having a dedicated log device, and it eliminates the
above-mentioned performance problems.
Log (ZIL) blocks can be allocated from the following locations. Each
one is tried in order until the allocation succeeds:
1. dedicated log vdevs, aka "slog" (spa_log_class)
2. embedded slog metaslabs (spa_embedded_log_class)
3. other metaslabs in normal vdevs (spa_normal_class)
The space required for the embedded slog metaslabs is usually between
0.5% and 1.0% of the pool, and comes out of the existing 3.2% of "slop"
space that is not available for user data.
On an all-ssd system with 4TB storage, 87% fragmentation, 60% capacity,
and recordsize=8k, testing shows a ~50% performance increase on random
8k sync writes. On even more fragmented systems (which hit problem #3
above and call txg_wait_synced()), the performance improvement can be
arbitrarily large (>100x).
Reviewed-by: Serapheim Dimitropoulos <serapheim@delphix.com>
Reviewed-by: George Wilson <gwilson@delphix.com>
Reviewed-by: Don Brady <don.brady@delphix.com>
Reviewed-by: Mark Maybee <mark.maybee@delphix.com>
Signed-off-by: Matthew Ahrens <mahrens@delphix.com>
Closes#11389
On a system with very high fragmentation, we may need to do lots of gang
allocations (e.g. most indirect block allocations (~50KB) may need to
gang). Before failing a "normal" allocation and resorting to ganging, we
try every metaslab. This has the impact of loading every metaslab (not
a huge deal since we now typically keep all metaslabs loaded), and also
iterating over every metaslab for every failing allocation. If there are
many metaslabs (more than the typical ~200, e.g. due to vdev expansion
or very large vdevs), the CPU cost of this iteration can be very
impactful. This iteration is done with the mg_lock held, creating long
hold times and high lock contention for concurrent allocations,
ultimately causing long txg sync times and poor application performance.
To address this, this commit changes the behavior of "normal" (not
try_hard, not ZIL) allocations. These will now only examine the 100
best metaslabs (as determined by their ms_weight). If none of these
have a large enough free segment, then the allocation will fail and
we'll fall back on ganging.
To accomplish this, we will now (normally) gang before doing a
`try_hard` allocation. Non-try_hard allocations will only examine the
100 best metaslabs of each vdev. In summary, we will first try normal
allocation. If that fails then we will do a gang allocation. If that
fails then we will do a "try hard" gang allocation. If that fails then
we will have a multi-layer gang block.
Reviewed-by: Paul Dagnelie <pcd@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Matthew Ahrens <mahrens@delphix.com>
Closes#11327
Investigating influence of scrub (especially sequential) on random read
latency I've noticed that on some HDDs single 4KB read may take up to 4
seconds! Deeper investigation shown that many HDDs heavily prioritize
sequential reads even when those are submitted with queue depth of 1.
This patch addresses the latency from two sides:
- by using _min_active queue depths for non-interactive requests while
the interactive request(s) are active and few requests after;
- by throttling it further if no interactive requests has completed
while configured amount of non-interactive did.
While there, I've also modified vdev_queue_class_to_issue() to give
more chances to schedule at least _min_active requests to the lowest
priorities. It should reduce starvation if several non-interactive
processes are running same time with some interactive and I think should
make possible setting of zfs_vdev_max_active to as low as 1.
I've benchmarked this change with 4KB random reads from ZVOL with 16KB
block size on newly written non-fragmented pool. On fragmented pool I
also saw improvements, but not so dramatic. Below are log2 histograms
of the random read latency in milliseconds for different devices:
4 2x mirror vdevs of SATA HDD WDC WD20EFRX-68EUZN0 before:
0, 0, 2, 1, 12, 21, 19, 18, 10, 15, 17, 21
after:
0, 0, 0, 24, 101, 195, 419, 250, 47, 4, 0, 0
, that means maximum latency reduction from 2s to 500ms.
4 2x mirror vdevs of SATA HDD WDC WD80EFZX-68UW8N0 before:
0, 0, 2, 31, 38, 28, 18, 12, 17, 20, 24, 10, 3
after:
0, 0, 55, 247, 455, 470, 412, 181, 36, 0, 0, 0, 0
, i.e. from 4s to 250ms.
1 SAS HDD SEAGATE ST14000NM0048 before:
0, 0, 29, 70, 107, 45, 27, 1, 0, 0, 1, 4, 19
after:
1, 29, 681, 1261, 676, 1633, 67, 1, 0, 0, 0, 0, 0
, i.e. from 4s to 125ms.
1 SAS SSD SEAGATE XS3840TE70014 before (microseconds):
0, 0, 0, 0, 0, 0, 0, 0, 70, 18343, 82548, 618
after:
0, 0, 0, 0, 0, 0, 0, 0, 283, 92351, 34844, 90
I've also measured scrub time during the test and on idle pools. On
idle fragmented pool I've measured scrub getting few percent faster
due to use of QD3 instead of QD2 before. On idle non-fragmented pool
I've measured no difference. On busy non-fragmented pool I've measured
scrub time increase about 1.5-1.7x, while IOPS increase reached 5-9x.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Signed-off-by: Alexander Motin <mav@FreeBSD.org>
Sponsored-By: iXsystems, Inc.
Closes#11166
The output of ZFS channel programs is logged on-disk in the zpool
history, and printed by `zpool history -i`. Channel programs can use
10MB of memory by default, and up to 100MB by using the `zfs program -m`
flag. Therefore their output can be up to some fraction of 100MB.
In addition to being somewhat wasteful of the limited space reserved for
the pool history (which for large pools is 1GB), in extreme cases this
can result in a failure of `ASSERT(length <= DMU_MAX_ACCESS);` in
`dmu_buf_hold_array_by_dnode()`.
This commit limits the output size that will be logged to 1MB. Larger
outputs will not be logged, instead a entry will be logged indicating
the size of the omitted output.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Matthew Ahrens <mahrens@delphix.com>
Closes#11194
This patch adds a new top-level vdev type called dRAID, which stands
for Distributed parity RAID. This pool configuration allows all dRAID
vdevs to participate when rebuilding to a distributed hot spare device.
This can substantially reduce the total time required to restore full
parity to pool with a failed device.
A dRAID pool can be created using the new top-level `draid` type.
Like `raidz`, the desired redundancy is specified after the type:
`draid[1,2,3]`. No additional information is required to create the
pool and reasonable default values will be chosen based on the number
of child vdevs in the dRAID vdev.
zpool create <pool> draid[1,2,3] <vdevs...>
Unlike raidz, additional optional dRAID configuration values can be
provided as part of the draid type as colon separated values. This
allows administrators to fully specify a layout for either performance
or capacity reasons. The supported options include:
zpool create <pool> \
draid[<parity>][:<data>d][:<children>c][:<spares>s] \
<vdevs...>
- draid[parity] - Parity level (default 1)
- draid[:<data>d] - Data devices per group (default 8)
- draid[:<children>c] - Expected number of child vdevs
- draid[:<spares>s] - Distributed hot spares (default 0)
Abbreviated example `zpool status` output for a 68 disk dRAID pool
with two distributed spares using special allocation classes.
```
pool: tank
state: ONLINE
config:
NAME STATE READ WRITE CKSUM
slag7 ONLINE 0 0 0
draid2:8d:68c:2s-0 ONLINE 0 0 0
L0 ONLINE 0 0 0
L1 ONLINE 0 0 0
...
U25 ONLINE 0 0 0
U26 ONLINE 0 0 0
spare-53 ONLINE 0 0 0
U27 ONLINE 0 0 0
draid2-0-0 ONLINE 0 0 0
U28 ONLINE 0 0 0
U29 ONLINE 0 0 0
...
U42 ONLINE 0 0 0
U43 ONLINE 0 0 0
special
mirror-1 ONLINE 0 0 0
L5 ONLINE 0 0 0
U5 ONLINE 0 0 0
mirror-2 ONLINE 0 0 0
L6 ONLINE 0 0 0
U6 ONLINE 0 0 0
spares
draid2-0-0 INUSE currently in use
draid2-0-1 AVAIL
```
When adding test coverage for the new dRAID vdev type the following
options were added to the ztest command. These options are leverages
by zloop.sh to test a wide range of dRAID configurations.
-K draid|raidz|random - kind of RAID to test
-D <value> - dRAID data drives per group
-S <value> - dRAID distributed hot spares
-R <value> - RAID parity (raidz or dRAID)
The zpool_create, zpool_import, redundancy, replacement and fault
test groups have all been updated provide test coverage for the
dRAID feature.
Co-authored-by: Isaac Huang <he.huang@intel.com>
Co-authored-by: Mark Maybee <mmaybee@cray.com>
Co-authored-by: Don Brady <don.brady@delphix.com>
Co-authored-by: Matthew Ahrens <mahrens@delphix.com>
Co-authored-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Mark Maybee <mmaybee@cray.com>
Reviewed-by: Matt Ahrens <matt@delphix.com>
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#10102
FreeBSD had this value tunable before the switch to the new OpenZFS.
The tunable name has changed, breaking legacy compat.
Restore legacy compat for this tunable, properly expose the tunable
with the new name on all platforms, and document it in
zfs-module-parameters(5).
While here, clean up the documentation for zfetch_max_distance a bit.
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes#11038
This change updates the documentation to refer to the project
as OpenZFS instead ZFS on Linux. Web links have been updated
to refer to https://github.com/openzfs/zfs. The extraneous
zfsonlinux.org web links in the ZED and SPL sources have been
dropped.
Reviewed-by: George Melikov <mail@gmelikov.ru>
Reviewed-by: Richard Laager <rlaager@wiktel.com>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#11007
with regard to evicted_l2_eligibile_mru. Even if l2arc_mfuonly is
enabled, this is not reflected in evicted_l2_eligible_mru as this
information is useful for deciding whether to toggle l2arc_mfuonly
depending on the current workload.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: George Amanakis <gamanakis@gmail.com>
Closes#10945
== Motivation and Context
The new vdev ashift optimization prevents the removal of devices when
a zfs configuration is comprised of disks which have different logical
and physical block sizes. This is caused because we set 'spa_min_ashift'
in vdev_open and then later call 'vdev_ashift_optimize'. This would
result in an inconsistency between spa's ashift calculations and that
of the top-level vdev.
In addition, the optimization logical ignores the overridden ashift
value that would be provided by '-o ashift=<val>'.
== Description
This change reworks the vdev ashift optimization so that it's only
set the first time the device is configured. It still allows the
physical and logical ahsift values to be set every time the device
is opened but those values are only consulted on first open.
Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Cedric Berger <cedric@precidata.com>
Signed-off-by: George Wilson <gwilson@delphix.com>
External-Issue: DLPX-71831
Closes#10932
Currently the ARC state (MFU/MRU) of cached L2ARC buffer and their
content type is unknown. Knowing this information may prove beneficial
in adjusting the L2ARC caching policy.
This commit adds L2ARC arcstats that display the aligned size
(in bytes) of L2ARC buffers according to their content type
(data/metadata) and according to their ARC state (MRU/MFU or
prefetch). It also expands the existing evict_l2_eligible arcstat to
differentiate between MFU and MRU buffers.
L2ARC caches buffers from the MRU and MFU lists of ARC. Upon caching a
buffer, its ARC state (MRU/MFU) is stored in the L2 header
(b_arcs_state). The l2_m{f,r}u_asize arcstats reflect the aligned size
(in bytes) of L2ARC buffers according to their ARC state (based on
b_arcs_state). We also account for the case where an L2ARC and ARC
cached MRU or MRU_ghost buffer transitions to MFU. The l2_prefetch_asize
reflects the alinged size (in bytes) of L2ARC buffers that were cached
while they had the prefetch flag set in ARC. This is dynamically updated
as the prefetch flag of L2ARC buffers changes.
When buffers are evicted from ARC, if they are determined to be L2ARC
eligible then their logical size is recorded in
evict_l2_eligible_m{r,f}u arcstats according to their ARC state upon
eviction.
Persistent L2ARC:
When committing an L2ARC buffer to a log block (L2ARC metadata) its
b_arcs_state and prefetch flag is also stored. If the buffer changes
its arcstate or prefetch flag this is reflected in the above arcstats.
However, the L2ARC metadata cannot currently be updated to reflect this
change.
Example: L2ARC caches an MRU buffer. L2ARC metadata and arcstats count
this as an MRU buffer. The buffer transitions to MFU. The arcstats are
updated to reflect this. Upon pool re-import or on/offlining the L2ARC
device the arcstats are cleared and the buffer will now be counted as an
MRU buffer, as the L2ARC metadata were not updated.
Bug fix:
- If l2arc_noprefetch is set, arc_read_done clears the L2CACHE flag of
an ARC buffer. However, prefetches may be issued in a way that
arc_read_done() is bypassed. Instead, move the related code in
l2arc_write_eligible() to account for those cases too.
Also add a test and update manpages for l2arc_mfuonly module parameter,
and update the manpages and code comments for l2arc_noprefetch.
Move persist_l2arc tests to l2arc.
Reviewed-by: Ryan Moeller <freqlabs@FreeBSD.org>
Reviewed-by: Richard Elling <Richard.Elling@RichardElling.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: George Amanakis <gamanakis@gmail.com>
Closes#10743
In certain workloads it may be beneficial to reduce wear of L2ARC
devices by not caching MRU metadata and data into L2ARC. This commit
introduces a new tunable l2arc_mfuonly for this purpose.
Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Reviewed-by: Richard Elling <Richard.Elling@RichardElling.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: George Amanakis <gamanakis@gmail.com>
Closes#10710
Duplicate io and checksum ereport events can misrepresent that
things are worse than they seem. Ideally the zpool events and the
corresponding vdev stat error counts in a zpool status should be
for unique errors -- not the same error being counted over and over.
This can be demonstrated in a simple example. With a single bad
block in a datafile and just 5 reads of the file we end up with a
degraded vdev, even though there is only one unique error in the pool.
The proposed solution to the above issue, is to eliminate duplicates
when posting events and when updating vdev error stats. We now save
recent error events of interest when posting events so that we can
easily check for duplicates when posting an error.
Reviewed by: Brad Lewis <brad.lewis@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Don Brady <don.brady@delphix.com>
Closes#10861
Since L2ARC buffers are not evicted on memory pressure, too large
amount of headers on system with irrationally large L2ARC can render
it slow or even unusable. This change limits L2ARC writes and
rebuild if unevictable L2ARC-only headers reach dangerous level.
While there, call arc_adapt() on L2ARC rebuild, so that it could
properly grow arc_c, reflecting potentially significant ARC size
increase and avoiding slow growth with hopeless eviction attempts
later when "overflow" is detected.
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reported-by: Richard Elling <Richard.Elling@RichardElling.com>
Signed-off-by: Alexander Motin <mav@FreeBSD.org>
Closes#10765
* Use all caps for document title.
* Remove section name as it can be inferred from the section number.
* Name "OpenZFS" as the document source.
* Bump modification date.
While here, fixed trailing whitespace reported by igor.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: George Melikov <mail@gmelikov.ru>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes#10792
Many modern devices use physical allocation units that are much
larger than the minimum logical allocation size accessible by
external commands. Two prevalent examples of this are 512e disk
drives (512b logical sector, 4K physical sector) and flash devices
(512b logical sector, 4K or larger allocation block size, and 128k
or larger erase block size). Operations that modify less than the
physical sector size result in a costly read-modify-write or garbage
collection sequence on these devices.
Simply exporting the true physical sector of the device to ZFS would
yield optimal performance, but has two serious drawbacks:
1. Existing pools created with devices that have different logical
and physical block sizes, but were configured to use the logical
block size (e.g. because the OS version used for pool construction
reported the logical block size instead of the physical block
size) will suddenly find that the vdev allocation size has
increased. This can be easily tolerated for active members of
the array, but ZFS would prevent replacement of a vdev with
another identical device because it now appears that the smaller
allocation size required by the pool is not supported by the new
device.
2. The device's physical block size may be too large to be supported
by ZFS. The optimal allocation size for the vdev may be quite
large. For example, a RAID controller may export a vdev that
requires read-modify-write cycles unless accessed using 64k
aligned/sized requests. ZFS currently has an 8k minimum block
size limit.
Reporting both the logical and physical allocation sizes for vdevs
solves these problems. A device may be used so long as the logical
block size is compatible with the configuration. By comparing the
logical and physical block sizes, new configurations can be optimized
and administrators can be notified of any existing pools that are
sub-optimal.
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Co-authored-by: Matthew Macy <mmacy@freebsd.org>
Signed-off-by: Matt Macy <mmacy@FreeBSD.org>
Closes#10619
We limit the size of nvlists passed to the kernel so a user cannot make
the kernel do an unreasonably large allocation. On FreeBSD this limit
was 128 kiB, which turns out to be a bit too small when doing some
operations involving a large number of datasets or snapshots, for
example replication.
Make this limit tunable, with a platform-specific auto default.
Linux keeps its limit at KMALLOC_MAX_SIZE. FreeBSD uses 1/4 of the
system limit on user wired memory, which allows it to scale depending
on system configuration.
Reviewed-by: Matt Macy <mmacy@FreeBSD.org>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ryan Moeller <freqlabs@FreeBSD.org>
Issue #6572Closes#10706
The ARC shrinker callback `arc_shrinker_count/_scan()` is invoked by the
kernel's shrinker mechanism when the system is running low on free
pages. This happens via 2 code paths:
1. "direct reclaim": The system is attempting to allocate a page, but we
are low on memory. The ARC shrinker callback is invoked from the
page-allocation code path.
2. "indirect reclaim": kswapd notices that there aren't many free pages,
so it invokes the ARC shrinker callback.
In both cases, the kernel's shrinker code requests that the ARC shrinker
callback release some of its cache, and then it measures how many pages
were released. However, it's measurement of released pages does not
include pages that are freed via `__free_pages()`, which is how the ARC
releases memory (via `abd_free_chunks()`). Rather, the kernel shrinker
code is looking for pages to be placed on the lists of reclaimable pages
(which is separate from actually-free pages).
Because the kernel shrinker code doesn't detect that the ARC has
released pages, it may call the ARC shrinker callback many times,
resulting in the ARC "collapsing" down to `arc_c_min`. This has several
negative impacts:
1. ZFS doesn't use RAM to cache data effectively.
2. In the direct reclaim case, a single page allocation may wait a long
time (e.g. more than a minute) while we evict the entire ARC.
3. Even with the improvements made in 67c0f0dedc ("ARC shrinking blocks
reads/writes"), occasionally `arc_size` may stay above `arc_c` for the
entire time of the ARC collapse, thus blocking ZFS read/write operations
in `arc_get_data_impl()`.
To address these issues, this commit limits the ways that the ARC
shrinker callback can be used by the kernel shrinker code, and mitigates
the impact of arc_is_overflowing() on ZFS read/write operations.
With this commit:
1. We limit the amount of data that can be reclaimed from the ARC via
the "direct reclaim" shrinker. This limits the amount of time it takes
to allocate a single page.
2. We do not allow the ARC to shrink via kswapd (indirect reclaim).
Instead we rely on `arc_evict_zthr` to monitor free memory and reduce
the ARC target size to keep sufficient free memory in the system. Note
that we can't simply rely on limiting the amount that we reclaim at once
(as for the direct reclaim case), because kswapd's "boosted" logic can
invoke the callback an unlimited number of times (see
`balance_pgdat()`).
3. When `arc_is_overflowing()` and we want to allocate memory,
`arc_get_data_impl()` will wait only for a multiple of the requested
amount of data to be evicted, rather than waiting for the ARC to no
longer be overflowing. This allows ZFS reads/writes to make progress
even while the ARC is overflowing, while also ensuring that the eviction
thread makes progress towards reducing the total amount of memory used
by the ARC.
4. The amount of memory that the ARC always tries to keep free for the
rest of the system, `arc_sys_free` is increased.
5. Now that the shrinker callback is able to provide feedback to the
kernel's shrinker code about our progress, we can safely enable
the kswapd hook. This will allow the arc to receive notifications
when memory pressure is first detected by the kernel. We also
re-enable the appropriate kstats to track these callbacks.
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Co-authored-by: George Wilson <george.wilson@delphix.com>
Signed-off-by: Matthew Ahrens <mahrens@delphix.com>
Closes#10600
Set the initial max sizes to ULONG_MAX to allow the caches to grow
with the ARC.
Recalculate the metadata cache size on demand so it can adapt, too.
Update descriptions in zfs-module-parameters(5).
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Matt Ahrens <matt@delphix.com>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes#10563Closes#10610
The device_rebuild feature enables sequential reconstruction when
resilvering. Mirror vdevs can be rebuilt in LBA order which may
more quickly restore redundancy depending on the pools average block
size, overall fragmentation and the performance characteristics
of the devices. However, block checksums cannot be verified
as part of the rebuild thus a scrub is automatically started after
the sequential resilver completes.
The new '-s' option has been added to the `zpool attach` and
`zpool replace` command to request sequential reconstruction
instead of healing reconstruction when resilvering.
zpool attach -s <pool> <existing vdev> <new vdev>
zpool replace -s <pool> <old vdev> <new vdev>
The `zpool status` output has been updated to report the progress
of sequential resilvering in the same way as healing resilvering.
The one notable difference is that multiple sequential resilvers
may be in progress as long as they're operating on different
top-level vdevs.
The `zpool wait -t resilver` command was extended to wait on
sequential resilvers. From this perspective they are no different
than healing resilvers.
Sequential resilvers cannot be supported for RAIDZ, but are
compatible with the dRAID feature being developed.
As part of this change the resilver_restart_* tests were moved
in to the functional/replacement directory. Additionally, the
replacement tests were renamed and extended to verify both
resilvering and rebuilding.
Original-patch-by: Isaac Huang <he.huang@intel.com>
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Reviewed-by: John Poduska <jpoduska@datto.com>
Co-authored-by: Mark Maybee <mmaybee@cray.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes#10349
Implement semi-compatible functionality for mode=0 (preallocation)
and mode=FALLOC_FL_KEEP_SIZE (preallocation beyond EOF) for ZPL.
Since ZFS does COW and snapshots, preallocating blocks for a file
cannot guarantee that writes to the file will not run out of space.
Even if the first overwrite was guaranteed, it would not handle any
later overwrite of blocks due to COW, so strict compliance is futile.
Instead, make a best-effort check that at least enough free space is
currently available in the pool (with a bit of margin), then create
a sparse file of the requested size and continue on with life.
This does not handle all cases (e.g. several fallocate() calls before
writing into the files when the filesystem is nearly full), which
would require a more complex mechanism to be implemented, probably
based on a modified version of dmu_prealloc(), but is usable as-is.
A new module option zfs_fallocate_reserve_percent is used to control
the reserve margin for any single fallocate call. By default, this
is 110% of the requested preallocation size, so an additional 10% of
available space is reserved for overhead to allow the application a
good chance of finishing the write when the fallocate() succeeds.
If the heuristics of this basic fallocate implementation are not
desirable, the old non-functional behavior of returning EOPNOTSUPP
for calls can be restored by setting zfs_fallocate_reserve_percent=0.
The parameter of zfs_statvfs() is changed to take an inode instead
of a dentry, since no dentry is available in zfs_fallocate_common().
A few tests from @behlendorf cover basic fallocate functionality.
Reviewed-by: Richard Laager <rlaager@wiktel.com>
Reviewed-by: Arshad Hussain <arshad.super@gmail.com>
Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Co-authored-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Andreas Dilger <adilger@dilger.ca>
Issue #326Closes#10408
Correct various typos in the comments and tests.
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Andrea Gelmini <andrea.gelmini@gelma.net>
Closes#10423
The l2arc_evict() function is responsible for evicting buffers which
reference the next bytes of the L2ARC device to be overwritten. Teach
this function to additionally TRIM that vdev space before it is
overwritten if the device has been filled with data. This is done by
vdev_trim_simple() which trims by issuing a new type of TRIM,
TRIM_TYPE_SIMPLE.
We also implement a "Trim Ahead" feature. It is a zfs module parameter,
expressed in % of the current write size. This trims ahead of the
current write size. A minimum of 64MB will be trimmed. The default is 0
which disables TRIM on L2ARC as it can put significant stress to
underlying storage devices. To enable TRIM on L2ARC we set
l2arc_trim_ahead > 0.
We also implement TRIM of the whole cache device upon addition to a
pool, pool creation or when the header of the device is invalid upon
importing a pool or onlining a cache device. This is dependent on
l2arc_trim_ahead > 0. TRIM of the whole device is done with
TRIM_TYPE_MANUAL so that its status can be monitored by zpool status -t.
We save the TRIM state for the whole device and the time of completion
on-disk in the header, and restore these upon L2ARC rebuild so that
zpool status -t can correctly report them. Whole device TRIM is done
asynchronously so that the user can export of the pool or remove the
cache device while it is trimming (ie if it is too slow).
We do not TRIM the whole device if persistent L2ARC has been disabled by
l2arc_rebuild_enabled = 0 because we may not want to lose all cached
buffers (eg we may want to import the pool with
l2arc_rebuild_enabled = 0 only once because of memory pressure). If
persistent L2ARC has been disabled by setting the module parameter
l2arc_rebuild_blocks_min_l2size to a value greater than the size of the
cache device then the whole device is trimmed upon creation or import of
a pool if l2arc_trim_ahead > 0.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Adam D. Moss <c@yotes.com>
Signed-off-by: George Amanakis <gamanakis@gmail.com>
Closes#9713Closes#9789Closes#10224
This commit makes the L2ARC persistent across reboots. We implement
a light-weight persistent L2ARC metadata structure that allows L2ARC
contents to be recovered after a reboot. This significantly eases the
impact a reboot has on read performance on systems with large caches.
Reviewed-by: Matthew Ahrens <mahrens@delphix.com>
Reviewed-by: George Wilson <gwilson@delphix.com>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Co-authored-by: Saso Kiselkov <skiselkov@gmail.com>
Co-authored-by: Jorgen Lundman <lundman@lundman.net>
Co-authored-by: George Amanakis <gamanakis@gmail.com>
Ported-by: Yuxuan Shui <yshuiv7@gmail.com>
Signed-off-by: George Amanakis <gamanakis@gmail.com>
Closes#925Closes#1823Closes#2672Closes#3744Closes#9582
Linux changed the default max ARC size to 1/2 of physical memory to
deal with shortcomings of the Linux SLUB allocator. Other platforms
do not require the same logic.
Implement an arc_default_max() function to determine a default max ARC
size in platform code.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Ryan Moeller <ryan@iXsystems.com>
Closes#10155
For each WRITE record in the stream, `zfs receive` creates a DMU
transaction (`dmu_tx_create()`) and writes this block's data into the
object. If per-block overheads (as opposed to per-byte overheads)
dominate performance (as is often the case with small recordsize), the
per-dmu-transaction overheads can be significant. For example, in some
workloads the `receieve_writer` thread is 100% on CPU, and more than
half of its CPU time is in these per-tx routines (e.g.
dmu_tx_hold_write, dmu_tx_assign, dmu_tx_commit).
To improve performance of `zfs receive`, this commit batches WRITE
records which are to nearby offsets of the same object, and uses one DMU
transaction to write them all. By default the batch size is 1MB, which
for recordsize=8K reduces the number of DMU transactions by 128x for
full send streams (incrementals will depend on how "clumpy" the changed
blocks are).
This commit improves the performance of `dd if=stream | zfs recv`
from 78,800 blocks/sec to 98,100 blocks/sec (25% improvement).
Reviewed-by: Paul Dagnelie <pcd@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Matthew Ahrens <mahrens@delphix.com>
Closes#10099
The module parameter zfs_async_block_max_blocks limits the number of
blocks that can be freed by the background freeing of filesystems and
snapshots (from "zfs destroy"), in one TXG. This is useful when freeing
dedup blocks, becuase each zio_free() of a dedup block can require an
i/o to read the relevant part of the dedup table (DDT), and will also
dirty that block.
zfs_async_block_max_blocks is set to 100,000 by default. For the more
typical case where dedup is not used, this can have a negative
performance impact on the rate of background freeing (from "zfs
destroy"). For example, with recordsize=8k, and TXG's syncing once
every 5 seconds, we can free only 160MB of data per second, which may be
much less than the rate we can write data.
This change increases zfs_async_block_max_blocks to be unlimited by
default. To address the dedup freeing issue, a new tunable is
introduced, zfs_max_async_dedup_frees, which limits the number of
zio_free()'s of dedup blocks done by background destroys, per txg. The
default is 100,000 free's (same as the old zfs_async_block_max_blocks
default).
Reviewed-by: Paul Dagnelie <pcd@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Matthew Ahrens <mahrens@delphix.com>
Closes#10000
Implements the RAID-Z function using AltiVec SIMD.
This is basically the NEON code translated to AltiVec.
Note that the 'fletcher' algorithm requires 64-bits
operations, and the initial implementations of AltiVec
(PPC74xx a.k.a. G4, PPC970 a.k.a. G5) only has up to
32-bits operations, so no 'fletcher'.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Romain Dolbeau <romain.dolbeau@european-processor-initiative.eu>
Closes#9539
I ran a report against the output of `modinfo zfs.ko`. This commit adds
everything missing and corrects a few renamed module parameters.
Specifically:
* zfs_checksums_per second renamed in ad796b8a3
* vdev_ms_count_limit renamed in c853f382d
Also fixes some variable type inconsistencies (unsigned int => uint)
Reviewed-by: George Amanakis <gamanakis@gmail.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: DHE <git@dehacked.net>
Closes#9809
As described in commit f81d5ef6 the zfs_vdev_elevator module
option is being removed. Users who require this functionality
should update their systems to set the disk scheduler using a
udev rule.
Reviewed-by: Richard Laager <rlaager@wiktel.com>
Reviewed-by: loli10K <ezomori.nozomu@gmail.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #8664Closes#9417Closes#9609
It is much faster than AVX512F when byteswapping on Skylake-SP
and newer, as we can do the byteswap in a single vshufb instead
of many instructions.
Reviewed by: Gvozden Neskovic <neskovic@gmail.com>
Reviewed-by: Chunwei Chen <tuxoko@gmail.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Romain Dolbeau <romain.dolbeau@atos.net>
Closes#9517
Originally the zfs_vdev_elevator module option was added as a
convenience so the requested elevator would be automatically set
on the underlying block devices. At the time this was simple
because the kernel provided an API function which did exactly this.
This API was then removed in the Linux 4.12 kernel which prompted
us to add compatibly code to set the elevator via a usermodehelper.
While well intentioned this introduced a bug which could cause a
system hang, that issue was subsequently fixed by commit 2a0d4188.
In order to avoid future bugs in this area, and to simplify the code,
this functionality is being deprecated. A console warning has been
added to notify any existing consumers and the documentation updated
accordingly. This option will remain for the lifetime of the 0.8.x
series for compatibility but if planned to be phased out of master.
Reviewed-by: Richard Laager <rlaager@wiktel.com>
Reviewed-by: loli10K <ezomori.nozomu@gmail.com>
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #8664Closes#9317
Currently the best way to wait for the completion of a long-running
operation in a pool, like a scrub or device removal, is to poll 'zpool
status' and parse its output, which is neither efficient nor convenient.
This change adds a 'wait' subcommand to the zpool command. When invoked,
'zpool wait' will block until a specified type of background activity
completes. Currently, this subcommand can wait for any of the following:
- Scrubs or resilvers to complete
- Devices to initialized
- Devices to be replaced
- Devices to be removed
- Checkpoints to be discarded
- Background freeing to complete
For example, a scrub that is in progress could be waited for by running
zpool wait -t scrub <pool>
This also adds a -w flag to the attach, checkpoint, initialize, replace,
remove, and scrub subcommands. When used, this flag makes the operations
kicked off by these subcommands synchronous instead of asynchronous.
This functionality is implemented using a new ioctl. The type of
activity to wait for is provided as input to the ioctl, and the ioctl
blocks until all activity of that type has completed. An ioctl was used
over other methods of kernel-userspace communiction primarily for the
sake of portability.
Porting Notes:
This is ported from Delphix OS change DLPX-44432. The following changes
were made while porting:
- Added ZoL-style ioctl input declaration.
- Reorganized error handling in zpool_initialize in libzfs to integrate
better with changes made for TRIM support.
- Fixed check for whether a checkpoint discard is in progress.
Previously it also waited if the pool had a checkpoint, instead of
just if a checkpoint was being discarded.
- Exposed zfs_initialize_chunk_size as a ZoL-style tunable.
- Updated more existing tests to make use of new 'zpool wait'
functionality, tests that don't exist in Delphix OS.
- Used existing ZoL tunable zfs_scan_suspend_progress, together with
zinject, in place of a new tunable zfs_scan_max_blks_per_txg.
- Added support for a non-integral interval argument to zpool wait.
Future work:
ZoL has support for trimming devices, which Delphix OS does not. In the
future, 'zpool wait' could be extended to add the ability to wait for
trim operations to complete.
Reviewed-by: Matt Ahrens <matt@delphix.com>
Reviewed-by: John Kennedy <john.kennedy@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: John Gallagher <john.gallagher@delphix.com>
Closes#9162
Reviewed-by: Ryan Moeller <ryan@ixsystems.com>
Reviewed-by: Richard Laager <rlaager@wiktel.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Andrea Gelmini <andrea.gelmini@gelma.net>
Closes#9233
With the other metaslab changes loaded onto a system, we can
significantly reduce the memory usage of each loaded metaslab and
unload them on demand if there is memory pressure. However, none
of those changes actually result in us keeping more metaslabs loaded.
If we don't keep more metaslabs loaded, we will still have to wait
for demand-loading to finish when no loaded metaslab can satisfy our
allocation, which can cause ZIL performance issues. In addition,
performance is traditionally measured by IOs per unit time, while
unloading is currently done on a txg-count basis. Txgs can take a
widely varying range of times, from tenths of a second to several
seconds. This can result in confusing, hard to predict behavior.
This change simply adds a time-based component to metaslab unloading.
A metaslab will remain loaded for one minute and 8 txgs (by default)
after it was last used, unless it is evicted due to memory pressure.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Matt Ahrens <mahrens@delphix.com>
Signed-off-by: Paul Dagnelie <pcd@delphix.com>
External-issue: DLPX-65016
External-issue: DLPX-65047
Closes#9197
On systems with large amounts of storage and high fragmentation, a huge
amount of space can be used by storing metaslab range trees. Since
metaslabs are only unloaded during a txg sync, and only if they have
been inactive for 8 txgs, it is possible to get into a state where all
of the system's memory is consumed by range trees and metaslabs, and
txgs cannot sync. While ZFS knows how to evict ARC data when needed,
it has no such mechanism for range tree data. This can result in boot
hangs for some system configurations.
First, we add the ability to unload metaslabs outside of syncing
context. Second, we store a multilist of all loaded metaslabs, sorted
by their selection txg, so we can quickly identify the oldest
metaslabs. We use a multilist to reduce lock contention during heavy
write workloads. Finally, we add logic that will unload a metaslab
when we're loading a new metaslab, if we're using more than a certain
fraction of the available memory on range trees.
Reviewed-by: Matt Ahrens <mahrens@delphix.com>
Reviewed-by: George Wilson <gwilson@delphix.com>
Reviewed-by: Sebastien Roy <sebastien.roy@delphix.com>
Reviewed-by: Serapheim Dimitropoulos <serapheim@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Paul Dagnelie <pcd@delphix.com>
Closes#9128
When a pool is imported it will scan the pool to verify the integrity
of the data and metadata. The amount it scans will depend on the
import flags provided. On systems with small amounts of memory or
when importing a pool from the crash kernel, it's possible for
spa_load_verify to issue too many I/Os that it consumes all the memory
of the system resulting in an OOM message or a hang.
To prevent this, we limit the amount of memory that the initial pool
scan can consume. This change will, by default, use 1/16th of the ARC
for scan I/Os to prevent running the system out of memory during import.
Reviewed-by: Matt Ahrens <matt@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Serapheim Dimitropoulos <serapheim@delphix.com>
Signed-off-by: George Wilson george.wilson@delphix.com
External-issue: DLPX-65237
External-issue: DLPX-65238
Closes#9146
When we unload metaslabs today in ZFS, the cached max_size value is
discarded. We instead use the histogram to determine whether or not we
think we can satisfy an allocation from the metaslab. This can result in
situations where, if we're doing I/Os of a size not aligned to a
histogram bucket, a metaslab is loaded even though it cannot satisfy the
allocation we think it can. For example, a metaslab with 16 entries in
the 16k-32k bucket may have entirely 16kB entries. If we try to allocate
a 24kB buffer, we will load that metaslab because we think it should be
able to handle the allocation. Doing so is expensive in CPU time, disk
reads, and average IO latency. This is exacerbated if the write being
attempted is a sync write.
This change makes ZFS cache the max_size after the metaslab is
unloaded. If we ever get a free (or a coalesced group of frees) larger
than the max_size, we will update it. Otherwise, we leave it as is. When
attempting to allocate, we use the max_size as a lower bound, and
respect it unless we are in try_hard. However, we do age the max_size
out at some point, since we expect the actual max_size to increase as we
do more frees. A more sophisticated algorithm here might be helpful, but
this works reasonably well.
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Matt Ahrens <mahrens@delphix.com>
Signed-off-by: Paul Dagnelie <pcd@delphix.com>
Closes#9055
Deleting a clone requires finding blocks are clone-only, not shared
with the snapshot. This was done by traversing the entire block tree
which results in a large performance penalty for sparsely
written clones.
This is new method keeps track of clone blocks when they are
modified in a "Livelist" so that, when it’s time to delete,
the clone-specific blocks are already at hand.
We see performance improvements because now deletion work is
proportional to the number of clone-modified blocks, not the size
of the original dataset.
Reviewed-by: Sean Eric Fagan <sef@ixsystems.com>
Reviewed-by: Matt Ahrens <matt@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Serapheim Dimitropoulos <serapheim@delphix.com>
Signed-off-by: Sara Hartse <sara.hartse@delphix.com>
Closes#8416
= Motivation
At Delphix we've seen a lot of customer systems where fragmentation
is over 75% and random writes take a performance hit because a lot
of time is spend on I/Os that update on-disk space accounting metadata.
Specifically, we seen cases where 20% to 40% of sync time is spend
after sync pass 1 and ~30% of the I/Os on the system is spent updating
spacemaps.
The problem is that these pools have existed long enough that we've
touched almost every metaslab at least once, and random writes
scatter frees across all metaslabs every TXG, thus appending to
their spacemaps and resulting in many I/Os. To give an example,
assuming that every VDEV has 200 metaslabs and our writes fit within
a single spacemap block (generally 4K) we have 200 I/Os. Then if we
assume 2 levels of indirection, we need 400 additional I/Os and
since we are talking about metadata for which we keep 2 extra copies
for redundancy we need to triple that number, leading to a total of
1800 I/Os per VDEV every TXG.
We could try and decrease the number of metaslabs so we have less
I/Os per TXG but then each metaslab would cover a wider range on
disk and thus would take more time to be loaded in memory from disk.
In addition, after it's loaded, it's range tree would consume more
memory.
Another idea would be to just increase the spacemap block size
which would allow us to fit more entries within an I/O block
resulting in fewer I/Os per metaslab and a speedup in loading time.
The problem is still that we don't deal with the number of I/Os
going up as the number of metaslabs is increasing and the fact
is that we generally write a lot to a few metaslabs and a little
to the rest of them. Thus, just increasing the block size would
actually waste bandwidth because we won't be utilizing our bigger
block size.
= About this patch
This patch introduces the Log Spacemap project which provides the
solution to the above problem while taking into account all the
aforementioned tradeoffs. The details on how it achieves that can
be found in the references sections below and in the code (see
Big Theory Statement in spa_log_spacemap.c).
Even though the change is fairly constraint within the metaslab
and lower-level SPA codepaths, there is a side-change that is
user-facing. The change is that VDEV IDs from VDEV holes will no
longer be reused. To give some background and reasoning for this,
when a log device is removed and its VDEV structure was replaced
with a hole (or was compacted; if at the end of the vdev array),
its vdev_id could be reused by devices added after that. Now
with the pool-wide space maps recording the vdev ID, this behavior
can cause problems (e.g. is this entry referring to a segment in
the new vdev or the removed log?). Thus, to simplify things the
ID reuse behavior is gone and now vdev IDs for top-level vdevs
are truly unique within a pool.
= Testing
The illumos implementation of this feature has been used internally
for a year and has been in production for ~6 months. For this patch
specifically there don't seem to be any regressions introduced to
ZTS and I have been running zloop for a week without any related
problems.
= Performance Analysis (Linux Specific)
All performance results and analysis for illumos can be found in
the links of the references. Redoing the same experiments in Linux
gave similar results. Below are the specifics of the Linux run.
After the pool reached stable state the percentage of the time
spent in pass 1 per TXG was 64% on average for the stock bits
while the log spacemap bits stayed at 95% during the experiment
(graph: sdimitro.github.io/img/linux-lsm/PercOfSyncInPassOne.png).
Sync times per TXG were 37.6 seconds on average for the stock
bits and 22.7 seconds for the log spacemap bits (related graph:
sdimitro.github.io/img/linux-lsm/SyncTimePerTXG.png). As a result
the log spacemap bits were able to push more TXGs, which is also
the reason why all graphs quantified per TXG have more entries for
the log spacemap bits.
Another interesting aspect in terms of txg syncs is that the stock
bits had 22% of their TXGs reach sync pass 7, 55% reach sync pass 8,
and 20% reach 9. The log space map bits reached sync pass 4 in 79%
of their TXGs, sync pass 7 in 19%, and sync pass 8 at 1%. This
emphasizes the fact that not only we spend less time on metadata
but we also iterate less times to convergence in spa_sync() dirtying
objects.
[related graphs:
stock- sdimitro.github.io/img/linux-lsm/NumberOfPassesPerTXGStock.png
lsm- sdimitro.github.io/img/linux-lsm/NumberOfPassesPerTXGLSM.png]
Finally, the improvement in IOPs that the userland gains from the
change is approximately 40%. There is a consistent win in IOPS as
you can see from the graphs below but the absolute amount of
improvement that the log spacemap gives varies within each minute
interval.
sdimitro.github.io/img/linux-lsm/StockVsLog3Days.png
sdimitro.github.io/img/linux-lsm/StockVsLog10Hours.png
= Porting to Other Platforms
For people that want to port this commit to other platforms below
is a list of ZoL commits that this patch depends on:
Make zdb results for checkpoint tests consistent
db587941c5
Update vdev_is_spacemap_addressable() for new spacemap encoding
419ba59145
Simplify spa_sync by breaking it up to smaller functions
8dc2197b7b
Factor metaslab_load_wait() in metaslab_load()
b194fab0fb
Rename range_tree_verify to range_tree_verify_not_present
df72b8bebe
Change target size of metaslabs from 256GB to 16GB
c853f382db
zdb -L should skip leak detection altogether
21e7cf5da8
vs_alloc can underflow in L2ARC vdevs
7558997d2f
Simplify log vdev removal code
6c926f426a
Get rid of space_map_update() for ms_synced_length
425d3237ee
Introduce auxiliary metaslab histograms
928e8ad47d
Error path in metaslab_load_impl() forgets to drop ms_sync_lock
8eef997679
= References
Background, Motivation, and Internals of the Feature
- OpenZFS 2017 Presentation:
youtu.be/jj2IxRkl5bQ
- Slides:
slideshare.net/SerapheimNikolaosDim/zfs-log-spacemaps-project
Flushing Algorithm Internals & Performance Results
(Illumos Specific)
- Blogpost:
sdimitro.github.io/post/zfs-lsm-flushing/
- OpenZFS 2018 Presentation:
youtu.be/x6D2dHRjkxw
- Slides:
slideshare.net/SerapheimNikolaosDim/zfs-log-spacemap-flushing-algorithm
Upstream Delphix Issues:
DLPX-51539, DLPX-59659, DLPX-57783, DLPX-61438, DLPX-41227, DLPX-59320
DLPX-63385
Reviewed-by: Sean Eric Fagan <sef@ixsystems.com>
Reviewed-by: Matt Ahrens <matt@delphix.com>
Reviewed-by: George Wilson <gwilson@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Serapheim Dimitropoulos <serapheim@delphix.com>
Closes#8442
Alexander Motin
наб
Rich Ercolani
Jitendra Patidar
Brian Behlendorf
Ryan Moeller
Andrea Gelmini
Don Brady
Matthew Ahrens
George Amanakis
George Wilson
Ryan Moeller
adilger
Matthew Ahrens
Romain Dolbeau
DeHackEd
Romain Dolbeau
John Gallagher
Paul Dagnelie
Sara Hartse
Serapheim Dimitropoulos