| 1 |
# SOME DESCRIPTIVE TITLE |
| 2 |
# Copyright (C) YEAR Free Software Foundation, Inc. |
| 3 |
# This file is distributed under the same license as the PACKAGE package. |
| 4 |
# FIRST AUTHOR <EMAIL@ADDRESS>, YEAR. |
| 5 |
# |
| 6 |
msgid "" |
| 7 |
msgstr "" |
| 8 |
"Project-Id-Version: PACKAGE VERSION\n" |
| 9 |
"POT-Creation-Date: 2019-01-14 13:37+0900\n" |
| 10 |
"PO-Revision-Date: YEAR-MO-DA HO:MI+ZONE\n" |
| 11 |
"Last-Translator: FULL NAME <EMAIL@ADDRESS>\n" |
| 12 |
"Language-Team: LANGUAGE <[email protected]>\n" |
| 13 |
"Language: \n" |
| 14 |
"MIME-Version: 1.0\n" |
| 15 |
"Content-Type: text/plain; charset=UTF-8\n" |
| 16 |
"Content-Transfer-Encoding: 8bit\n" |
| 17 |
|
| 18 |
#. type: Title == |
| 19 |
#, no-wrap |
| 20 |
msgid "Measuring block device performance" |
| 21 |
msgstr "ブロックデバイスのパフォーマンス測定" |
| 22 |
|
| 23 |
#. type: Title === |
| 24 |
#, no-wrap |
| 25 |
msgid "Measuring throughput" |
| 26 |
msgstr "スループットの測定" |
| 27 |
|
| 28 |
#. type: Plain text |
| 29 |
msgid "" |
| 30 |
"When measuring the impact of using DRBD on a system's I/O throughput, the " |
| 31 |
"_absolute_ throughput the system is capable of is of little relevance. What " |
| 32 |
"is much more interesting is the _relative_ impact DRBD has on I/O " |
| 33 |
"performance. Thus it is always necessary to measure I/O throughput both with " |
| 34 |
"and without DRBD." |
| 35 |
msgstr "" |
| 36 |
"DRBDを使用することによるシステムのI/Oスループットへの影響を測定する際には、ス" |
| 37 |
"ループットの _絶対値_ はほとんど意味がありません。必要なのは、DRBDがI/Oパ" |
| 38 |
"フォーマンスに及ぼす _相対的_ な影響です。したがって、I/Oスループットを測定す" |
| 39 |
"る際には、必ずDRBDがある場合とない場合の両方を測定する必要があります。" |
| 40 |
|
| 41 |
#. type: Plain text |
| 42 |
msgid "" |
| 43 |
"The tests described in this section are intrusive; they overwrite data and " |
| 44 |
"bring DRBD devices out of sync. It is thus vital that you perform them only " |
| 45 |
"on scratch volumes which can be discarded after testing has completed." |
| 46 |
msgstr "" |
| 47 |
"ここで説明するテストは過激なものです。データが上書きされるため、DRBDデバイス" |
| 48 |
"の同期が失われます。そのためこのテストは、テストが完了したら破棄できるスク" |
| 49 |
"ラッチボリュームで行ってください。" |
| 50 |
|
| 51 |
#. type: Plain text |
| 52 |
msgid "" |
| 53 |
"I/O throughput estimation works by writing significantly large chunks of " |
| 54 |
"data to a block device, and measuring the amount of time the system took to " |
| 55 |
"complete the write operation. This can be easily done using a fairly " |
| 56 |
"ubiquitous utility, `dd`, whose reasonably recent versions include a built-" |
| 57 |
"in throughput estimation." |
| 58 |
msgstr "" |
| 59 |
"I/Oスループットを見積もるには、比較的大きなデータチャンクをブロックデバイスに" |
| 60 |
"書き込み、システムが書き込み操作を完了するまでの時間を測定します。これは一般" |
| 61 |
"的なユーティリティである `dd` を使用して簡単に測定できます。比較的新しいバー" |
| 62 |
"ジョンにはスループット見積もり機能が組み込まれています。" |
| 63 |
|
| 64 |
#. type: Plain text |
| 65 |
msgid "" |
| 66 |
"A simple ``dd``-based throughput benchmark, assuming you have a scratch " |
| 67 |
"resource named `test`, which is currently connected and in the secondary " |
| 68 |
"role on both nodes, is one like the following:" |
| 69 |
msgstr "" |
| 70 |
"簡単な `dd` ベースのスループットベンチマークは、たとえば次のようになります。" |
| 71 |
"ここでは、test という名前のスクラッチリソースが現在接続されており、両方のノー" |
| 72 |
"ドでセカンダリロールになっています。" |
| 73 |
|
| 74 |
#. type: delimited block - |
| 75 |
#, no-wrap |
| 76 |
msgid "" |
| 77 |
"# TEST_RESOURCE=test\n" |
| 78 |
"# TEST_DEVICE=$(drbdadm sh-dev $TEST_RESOURCE | head -1)\n" |
| 79 |
"# TEST_LL_DEVICE=$(drbdadm sh-ll-dev $TEST_RESOURCE | head -1)\n" |
| 80 |
"# drbdadm primary $TEST_RESOURCE\n" |
| 81 |
"# for i in $(seq 5); do\n" |
| 82 |
" dd if=/dev/zero of=$TEST_DEVICE bs=1M count=512 oflag=direct\n" |
| 83 |
" done\n" |
| 84 |
"# drbdadm down $TEST_RESOURCE\n" |
| 85 |
"# for i in $(seq 5); do\n" |
| 86 |
" dd if=/dev/zero of=$TEST_LL_DEVICE bs=1M count=512 oflag=direct\n" |
| 87 |
" done\n" |
| 88 |
msgstr "" |
| 89 |
"# TEST_RESOURCE=test\n" |
| 90 |
"# TEST_DEVICE=$(drbdadm sh-dev $TEST_RESOURCE | head -1)\n" |
| 91 |
"# TEST_LL_DEVICE=$(drbdadm sh-ll-dev $TEST_RESOURCE | head -1)\n" |
| 92 |
"# drbdadm primary $TEST_RESOURCE\n" |
| 93 |
"# for i in $(seq 5); do\n" |
| 94 |
" dd if=/dev/zero of=$TEST_DEVICE bs=1M count=512 oflag=direct\n" |
| 95 |
" done\n" |
| 96 |
"# drbdadm down $TEST_RESOURCE\n" |
| 97 |
"# for i in $(seq 5); do\n" |
| 98 |
" dd if=/dev/zero of=$TEST_LL_DEVICE bs=1M count=512 oflag=direct\n" |
| 99 |
" done\n" |
| 100 |
|
| 101 |
#. type: Plain text |
| 102 |
msgid "" |
| 103 |
"This test simply writes 512MiB of data to your DRBD device, and then to its " |
| 104 |
"backing device for comparison. Both tests are repeated 5 times each to allow " |
| 105 |
"for some statistical averaging. The relevant result is the throughput " |
| 106 |
"measurements generated by `dd`." |
| 107 |
msgstr "" |
| 108 |
"このテストでは、512MのデータチャンクをDRBDデバイスに書き込み、次に比較のため" |
| 109 |
"に下位デバイスに書き込みます。両方のテストをそれぞれ5回繰り返し、統計平均を求" |
| 110 |
"めます。この結果は `dd` が生成したスループット測定値です。" |
| 111 |
|
| 112 |
#. type: Plain text |
| 113 |
msgid "" |
| 114 |
"For freshly enabled DRBD devices, it is normal to see slightly reduced " |
| 115 |
"performance on the first `dd` run. This is due to the Activity Log being " |
| 116 |
"\"cold\", and is no cause for concern." |
| 117 |
msgstr "" |
| 118 |
"新規に有効にしたDRBDデバイスの場合は、最初の `dd` 実行ではパフォーマンスがか" |
| 119 |
"なり低くなりますが、これは正常な状態です。アクティビティログがいわゆる「コー" |
| 120 |
"ルド」な状態のためで、問題はありません。" |
| 121 |
|
| 122 |
#. type: Plain text |
| 123 |
msgid "See our <<ch-throughput>> chapter for some performance numbers." |
| 124 |
msgstr "パフォーマンスの参考値については<<ch-throughput>>をご参照ください。" |
| 125 |
|
| 126 |
#. type: Title === |
| 127 |
#, no-wrap |
| 128 |
msgid "Measuring latency" |
| 129 |
msgstr "レイテンシの測定" |
| 130 |
|
| 131 |
#. type: Plain text |
| 132 |
msgid "" |
| 133 |
"Latency measurements have objectives completely different from throughput " |
| 134 |
"benchmarks: in I/O latency tests, one writes a very small chunk of data " |
| 135 |
"(ideally the smallest chunk of data that the system can deal with), and " |
| 136 |
"observes the time it takes to complete that write. The process is usually " |
| 137 |
"repeated several times to account for normal statistical fluctuations." |
| 138 |
msgstr "" |
| 139 |
"レイテンシ測定はスループット測定とはまったく異なります。I/Oレイテンシテストで" |
| 140 |
"は非常に小さいデータチャンク(理想的にはシステムが処理可能な最も小さいデータ" |
| 141 |
"チャンク)を書き込み、書き込みが完了するまでの時間を測定します。通常はこれを何" |
| 142 |
"度か繰り返して、通常の統計変動を相殺します。" |
| 143 |
|
| 144 |
#. type: Plain text |
| 145 |
msgid "" |
| 146 |
"Just as throughput measurements, I/O latency measurements may be performed " |
| 147 |
"using the ubiquitous `dd` utility, albeit with different settings and an " |
| 148 |
"entirely different focus of observation." |
| 149 |
msgstr "" |
| 150 |
"スループットの測定と同様に、I/O待ち時間の測定にも一般的な `dd` ユーティリティ" |
| 151 |
"を使用できますが、異なる設定とまったく異なった測定を行います。" |
| 152 |
|
| 153 |
#. type: Plain text |
| 154 |
msgid "" |
| 155 |
"Provided below is a simple ``dd``-based latency micro-benchmark, assuming " |
| 156 |
"you have a scratch resource named `test` which is currently connected and in " |
| 157 |
"the secondary role on both nodes:" |
| 158 |
msgstr "" |
| 159 |
"以下は簡単な `dd` ベースの待ち時間のマイクロベンチマークです。ここでは、 " |
| 160 |
"`test` という名前のスクラッチリソースが現在接続されており、両方のノードでセカ" |
| 161 |
"ンダリロールになっています。" |
| 162 |
|
| 163 |
#. type: delimited block - |
| 164 |
#, no-wrap |
| 165 |
msgid "" |
| 166 |
"# TEST_RESOURCE=test\n" |
| 167 |
"# TEST_DEVICE=$(drbdadm sh-dev $TEST_RESOURCE | head -1)\n" |
| 168 |
"# TEST_LL_DEVICE=$(drbdadm sh-ll-dev $TEST_RESOURCE | head -1)\n" |
| 169 |
"# drbdadm primary $TEST_RESOURCE\n" |
| 170 |
"# dd if=/dev/zero of=$TEST_DEVICE bs=4k count=1000 oflag=direct\n" |
| 171 |
"# drbdadm down $TEST_RESOURCE\n" |
| 172 |
"# dd if=/dev/zero of=$TEST_LL_DEVICE bs=4k count=1000 oflag=direct\n" |
| 173 |
msgstr "" |
| 174 |
"# TEST_RESOURCE=test\n" |
| 175 |
"# TEST_DEVICE=$(drbdadm sh-dev $TEST_RESOURCE | head -1)\n" |
| 176 |
"# TEST_LL_DEVICE=$(drbdadm sh-ll-dev $TEST_RESOURCE | head -1)\n" |
| 177 |
"# drbdadm primary $TEST_RESOURCE\n" |
| 178 |
"# dd if=/dev/zero of=$TEST_DEVICE bs=4k count=1000 oflag=direct\n" |
| 179 |
"# drbdadm down $TEST_RESOURCE\n" |
| 180 |
"# dd if=/dev/zero of=$TEST_LL_DEVICE bs=4k count=1000 oflag=direct\n" |
| 181 |
|
| 182 |
#. type: Plain text |
| 183 |
msgid "" |
| 184 |
"This test writes 1,000 chunks with 4kiB each to your DRBD device, and then " |
| 185 |
"to its backing device for comparison. 4096 bytes is the smallest block size " |
| 186 |
"that a Linux system (on all architectures except s390), modern hard disks, " |
| 187 |
"and SSDs, are expected to handle." |
| 188 |
msgstr "" |
| 189 |
"このテストでは、4kiBのデータチャンクをDRBDデバイスに1,000回書き込み、次に比" |
| 190 |
"較のために下位デバイスにも1000回書き込みます。4096バイトはLinuxシステム(s390" |
| 191 |
"以外のすべてのアーキテクチャ)が扱うことができる最小のブロックサイズです。" |
| 192 |
|
| 193 |
#. type: Plain text |
| 194 |
msgid "" |
| 195 |
"It is important to understand that throughput measurements generated by `dd` " |
| 196 |
"are completely irrelevant for this test; what is important is the _time_ " |
| 197 |
"elapsed during the completion of said 1,000 writes. Dividing this time by " |
| 198 |
"1,000 gives the average latency of a single block write." |
| 199 |
msgstr "" |
| 200 |
"`dd` が生成するスループット測定はこのテストではまったく意味がなく、ここで重要" |
| 201 |
"なのは、1,000回の書き込みが完了するまでにかかった時間です。この時間を1,000で" |
| 202 |
"割ると、1つのセクタへの書き込みの平均待ち時間を求めることができます。" |
| 203 |
|
| 204 |
#. type: Plain text |
| 205 |
msgid "" |
| 206 |
"This is the _worst-case_, in that it is single-threaded and does one write " |
| 207 |
"strictly after the one before, i.e. runs with an I/O-depth of 1. Please take " |
| 208 |
"a look at <<s-latency-iops>>." |
| 209 |
msgstr "" |
| 210 |
"これは、シングルスレッドで厳密に1つずつ書き込みをしますので、最悪のケースで" |
| 211 |
"はあります。つまり、I/O-depthが1の場合などです。<<s-latency-iops>>をご参照く" |
| 212 |
"ださい。" |
| 213 |
|
| 214 |
#. type: Plain text |
| 215 |
msgid "" |
| 216 |
"Furthermore, see our <<ch-latency>> chapter for some typical performance " |
| 217 |
"values." |
| 218 |
msgstr "参考情報として<<ch-latency>>もご参照ください。" |