mirror_zfs/cmd
Matthew Ahrens 85ec5cbae2
Include scatter_chunk_waste in arc_size
The ARC caches data in scatter ABD's, which are collections of pages,
which are typically 4K.  Therefore, the space used to cache each block
is rounded up to a multiple of 4K.  The ABD subsystem tracks this wasted
memory in the `scatter_chunk_waste` kstat.  However, the ARC's `size` is
not aware of the memory used by this round-up, it only accounts for the
size that it requested from the ABD subsystem.

Therefore, the ARC is effectively using more memory than it is aware of,
due to the `scatter_chunk_waste`.  This impacts observability, e.g.
`arcstat` will show that the ARC is using less memory than it
effectively is.  It also impacts how the ARC responds to memory
pressure.  As the amount of `scatter_chunk_waste` changes, it appears to
the ARC as memory pressure, so it needs to resize `arc_c`.

If the sector size (`1<<ashift`) is the same as the page size (or
larger), there won't be any waste.  If the (compressed) block size is
relatively large compared to the page size, the amount of
`scatter_chunk_waste` will be small, so the problematic effects are
minimal.

However, if using 512B sectors (`ashift=9`), and the (compressed) block
size is small (e.g. `compression=on` with the default `volblocksize=8k`
or a decreased `recordsize`), the amount of `scatter_chunk_waste` can be
very large.  On a production system, with `arc_size` at a constant 50%
of memory, `scatter_chunk_waste` has been been observed to be 10-30% of
memory.

This commit adds `scatter_chunk_waste` to `arc_size`, and adds a new
`waste` field to `arcstat`.  As a result, the ARC's memory usage is more
observable, and `arc_c` does not need to be adjusted as frequently.

Reviewed-by: Pavel Zakharov <pavel.zakharov@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: George Wilson <gwilson@delphix.com>
Reviewed-by: Ryan Moeller <ryan@iXsystems.com>
Signed-off-by: Matthew Ahrens <mahrens@delphix.com>
Closes #10701
2020-08-17 20:04:04 -07:00
..
arc_summary Fix reporting of L2ARC writes in arc_summary3 2020-08-17 11:04:06 -07:00
arcstat Include scatter_chunk_waste in arc_size 2020-08-17 20:04:04 -07:00
dbufstat Centralize variable substitution 2020-07-14 17:33:44 -07:00
fsck_zfs Fix typos in cmd/ 2019-08-30 09:43:30 -07:00
mount_zfs Use abs_top_builddir when referencing libraries 2020-07-10 14:26:32 -07:00
raidz_test Unconditionally enable debugging for libzpool 2020-07-10 15:30:31 -07:00
vdev_id Fix typos in cmd/ 2019-08-30 09:43:30 -07:00
zdb FreeBSD: Add legacy arc_min and arc_max 2020-07-19 10:15:34 -07:00
zed Centralize variable substitution 2020-07-14 17:33:44 -07:00
zfs 'zfs share -a' should handle 'canmount=noauto' 2020-08-11 13:55:04 -07:00
zfs_ids_to_path Use abs_top_builddir when referencing libraries 2020-07-10 14:26:32 -07:00
zgenhostid bash scripts: use /usr/bin/env for bash shebangs 2020-02-10 13:13:46 -08:00
zhack Unconditionally enable debugging for libzpool 2020-07-10 15:30:31 -07:00
zinject Use abs_top_builddir when referencing libraries 2020-07-10 14:26:32 -07:00
zpool Changes to make openzfs build within FreeBSD buildworld 2020-07-31 21:30:31 -07:00
zstream Add support to decode a resume token 2020-07-23 17:44:03 -07:00
zstreamdump Minor zstream redup command fixes 2020-04-10 21:10:09 -07:00
ztest Rename refcount.h to zfs_refcount.h 2020-07-29 16:35:33 -07:00
zvol_id OpenZFS restructuring - libspl 2019-10-02 10:39:48 -07:00
zvol_wait zvol_wait should ignore redacted zvols 2019-11-06 10:51:19 -08:00
Makefile.am Small program that converts a dataset id and an object id to a path 2020-05-20 10:05:33 -07:00