docs/qcow2-cache.txt

   1 qcow2 L2/refcount cache configuration
   2 =====================================
   3 Copyright (C) 2015, 2018 Igalia, S.L.
   4 Author: Alberto Garcia <berto@igalia.com>
   5
   6 This work is licensed under the terms of the GNU GPL, version 2 or
   7 later. See the COPYING file in the top-level directory.
   8
   9 Introduction
  10 ------------
  11 The QEMU qcow2 driver has two caches that can improve the I/O
  12 performance significantly. However, setting the right cache sizes is
  13 not a straightforward operation.
  14
  15 This document attempts to give an overview of the L2 and refcount
  16 caches, and how to configure them.
  17
  18 Please refer to the docs/interop/qcow2.txt file for an in-depth
  19 technical description of the qcow2 file format.
  20
  21
  22 Clusters
  23 --------
  24 A qcow2 file is organized in units of constant size called clusters.
  25
  26 The cluster size is configurable, but it must be a power of two and
  27 its value 512 bytes or higher. QEMU currently defaults to 64 KB
  28 clusters, and it does not support sizes larger than 2MB.
  29
  30 The 'qemu-img create' command supports specifying the size using the
  31 cluster_size option:
  32
  33    qemu-img create -f qcow2 -o cluster_size=128K hd.qcow2 4G
  34
  35
  36 The L2 tables
  37 -------------
  38 The qcow2 format uses a two-level structure to map the virtual disk as
  39 seen by the guest to the disk image in the host. These structures are
  40 called the L1 and L2 tables.
  41
  42 There is one single L1 table per disk image. The table is small and is
  43 always kept in memory.
  44
  45 There can be many L2 tables, depending on how much space has been
  46 allocated in the image. Each table is one cluster in size. In order to
  47 read or write data from the virtual disk, QEMU needs to read its
  48 corresponding L2 table to find out where that data is located. Since
  49 reading the table for each I/O operation can be expensive, QEMU keeps
  50 an L2 cache in memory to speed up disk access.
  51
  52 The size of the L2 cache can be configured, and setting the right
  53 value can improve the I/O performance significantly.
  54
  55
  56 The refcount blocks
  57 -------------------
  58 The qcow2 format also mantains a reference count for each cluster.
  59 Reference counts are used for cluster allocation and internal
  60 snapshots. The data is stored in a two-level structure similar to the
  61 L1/L2 tables described above.
  62
  63 The second level structures are called refcount blocks, are also one
  64 cluster in size and the number is also variable and dependent on the
  65 amount of allocated space.
  66
  67 Each block contains a number of refcount entries. Their size (in bits)
  68 is a power of two and must not be higher than 64. It defaults to 16
  69 bits, but a different value can be set using the refcount_bits option:
  70
  71    qemu-img create -f qcow2 -o refcount_bits=8 hd.qcow2 4G
  72
  73 QEMU keeps a refcount cache to speed up I/O much like the
  74 aforementioned L2 cache, and its size can also be configured.
  75
  76
  77 Choosing the right cache sizes
  78 ------------------------------
  79 In order to choose the cache sizes we need to know how they relate to
  80 the amount of allocated space.
  81
  82 The amount of virtual disk that can be mapped by the L2 and refcount
  83 caches (in bytes) is:
  84
  85    disk_size = l2_cache_size * cluster_size / 8
  86    disk_size = refcount_cache_size * cluster_size * 8 / refcount_bits
  87
  88 With the default values for cluster_size (64KB) and refcount_bits
  89 (16), that is
  90
  91    disk_size = l2_cache_size * 8192
  92    disk_size = refcount_cache_size * 32768
  93
  94 So in order to cover n GB of disk space with the default values we
  95 need:
  96
  97    l2_cache_size = disk_size_GB * 131072
  98    refcount_cache_size = disk_size_GB * 32768
  99
 100 QEMU has a default L2 cache of 1MB (1048576 bytes) and a refcount
 101 cache of 256KB (262144 bytes), so using the formulas we've just seen
 102 we have
 103
 104    1048576 / 131072 = 8 GB of virtual disk covered by that cache
 105     262144 /  32768 = 8 GB
 106
 107
 108 How to configure the cache sizes
 109 --------------------------------
 110 Cache sizes can be configured using the -drive option in the
 111 command-line, or the 'blockdev-add' QMP command.
 112
 113 There are three options available, and all of them take bytes:
 114
 115 "l2-cache-size":         maximum size of the L2 table cache
 116 "refcount-cache-size":   maximum size of the refcount block cache
 117 "cache-size":            maximum size of both caches combined
 118
 119 There are a few things that need to be taken into account:
 120
 121  - Both caches must have a size that is a multiple of the cluster size
 122    (or the cache entry size: see "Using smaller cache sizes" below).
 123
 124  - The default L2 cache size is 8 clusters or 1MB (whichever is more),
 125    and the minimum is 2 clusters (or 2 cache entries, see below).
 126
 127  - The default (and minimum) refcount cache size is 4 clusters.
 128
 129  - If only "cache-size" is specified then QEMU will assign as much
 130    memory as possible to the L2 cache before increasing the refcount
 131    cache size.
 132
 133 Unlike L2 tables, refcount blocks are not used during normal I/O but
 134 only during allocations and internal snapshots. In most cases they are
 135 accessed sequentially (even during random guest I/O) so increasing the
 136 refcount cache size won't have any measurable effect in performance
 137 (this can change if you are using internal snapshots, so you may want
 138 to think about increasing the cache size if you use them heavily).
 139
 140 Before QEMU 2.12 the refcount cache had a default size of 1/4 of the
 141 L2 cache size. This resulted in unnecessarily large caches, so now the
 142 refcount cache is as small as possible unless overridden by the user.
 143
 144
 145 Using smaller cache entries
 146 ---------------------------
 147 The qcow2 L2 cache stores complete tables by default. This means that
 148 if QEMU needs an entry from an L2 table then the whole table is read
 149 from disk and is kept in the cache. If the cache is full then a
 150 complete table needs to be evicted first.
 151
 152 This can be inefficient with large cluster sizes since it results in
 153 more disk I/O and wastes more cache memory.
 154
 155 Since QEMU 2.12 you can change the size of the L2 cache entry and make
 156 it smaller than the cluster size. This can be configured using the
 157 "l2-cache-entry-size" parameter:
 158
 159    -drive file=hd.qcow2,l2-cache-size=2097152,l2-cache-entry-size=4096
 160
 161 Some things to take into account:
 162
 163  - The L2 cache entry size has the same restrictions as the cluster
 164    size (power of two, at least 512 bytes).
 165
 166  - Smaller entry sizes generally improve the cache efficiency and make
 167    disk I/O faster. This is particularly true with solid state drives
 168    so it's a good idea to reduce the entry size in those cases. With
 169    rotating hard drives the situation is a bit more complicated so you
 170    should test it first and stay with the default size if unsure.
 171
 172  - Try different entry sizes to see which one gives faster performance
 173    in your case. The block size of the host filesystem is generally a
 174    good default (usually 4096 bytes in the case of ext4).
 175
 176  - Only the L2 cache can be configured this way. The refcount cache
 177    always uses the cluster size as the entry size.
 178
 179  - If the L2 cache is big enough to hold all of the image's L2 tables
 180    (as explained in the "Choosing the right cache sizes" section
 181    earlier in this document) then none of this is necessary and you
 182    can omit the "l2-cache-entry-size" parameter altogether.
 183
 184
 185 Reducing the memory usage
 186 -------------------------
 187 It is possible to clean unused cache entries in order to reduce the
 188 memory usage during periods of low I/O activity.
 189
 190 The parameter "cache-clean-interval" defines an interval (in seconds).
 191 All cache entries that haven't been accessed during that interval are
 192 removed from memory.
 193
 194 This example removes all unused cache entries every 15 minutes:
 195
 196    -drive file=hd.qcow2,cache-clean-interval=900
 197
 198 If unset, the default value for this parameter is 0 and it disables
 199 this feature.
 200
 201 Note that this functionality currently relies on the MADV_DONTNEED
 202 argument for madvise() to actually free the memory. This is a
 203 Linux-specific feature, so cache-clean-interval is not supported in
 204 other systems.