# SOME DESCRIPTIVE TITLE # Copyright (C) YEAR Free Software Foundation, Inc. # This file is distributed under the same license as the PACKAGE package. # FIRST AUTHOR , YEAR. # #, fuzzy msgid "" msgstr "" "Project-Id-Version: PACKAGE VERSION\n" "POT-Creation-Date: 2020-03-02 15:09+0000\n" "PO-Revision-Date: YEAR-MO-DA HO:MI+ZONE\n" "Last-Translator: FULL NAME \n" "Language-Team: LANGUAGE \n" "Language: \n" "MIME-Version: 1.0\n" "Content-Type: text/plain; charset=CHARSET\n" "Content-Transfer-Encoding: 8bit\n" #. type: Title == #: throughput.adoc:2 #, no-wrap msgid "Optimizing DRBD throughput" msgstr "优化DRBD吞吐量" #. type: Plain text #: throughput.adoc:7 msgid "" "This chapter deals with optimizing DRBD throughput. It examines some " "hardware considerations with regard to throughput optimization, and details " "tuning recommendations for that purpose." msgstr "本章讨论优化DRBD吞吐量。它研究了与吞吐量优化有关的一些硬件考虑因素,并详细介绍了为此目的提出的优化建议。" #. type: Title === #: throughput.adoc:9 #, no-wrap msgid "Hardware considerations" msgstr "硬件注意事项" #. type: Plain text #: throughput.adoc:14 msgid "" "DRBD throughput is affected by both the bandwidth of the underlying I/O " "subsystem (disks, controllers, and corresponding caches), and the bandwidth " "of the replication network." msgstr "DRBD吞吐量受底层I/O子系统(磁盘、控制器和相应缓存)的带宽和复制网络的带宽的影响。" #. type: Block title #: throughput.adoc:15 #, no-wrap msgid "I/O subsystem throughput" msgstr "I/O子系统吞吐量" #. type: Plain text #: throughput.adoc:28 msgid "" "indexterm:[throughput]I/O subsystem throughput is determined, largely, by " "the number and type of storage units (disks, SSDs, other Flash storage [like" " FusionIO], ...) that can be written to in parallel. A single, reasonably " "recent, SCSI or SAS disk will typically allow streaming writes of roughly " "40MiB/s to the single disk; an SSD will do 300MiB/s; one of the recent Flash" " storages (NVMe) will be at 1GiB/s. When deployed in a striping " "configuration, the I/O subsystem will parallelize writes across disks, " "effectively multiplying a single disk's throughput by the number of stripes " "in the configuration. Thus the same, 40MB/s disks will allow effective " "throughput of 120MB/s in a RAID-0 or RAID-1+0 configuration with three " "stripes, or 200MB/s with five stripes; with SSDs and/or NVMe you can easily " "get to 1GiB/sec." msgstr "" "indexterm:[throughput]I/O子系统吞吐量在很大程度上取决于可并行写入的存储单元(磁盘、固态硬盘、其他闪存[如FusionIO],...)的数量和类型。一个相当新的SCSI或SAS磁盘通常允许向单个磁盘流式写入大约40MiB/秒;一个固态硬盘将执行300MiB/秒;一个最新的闪存(NVMe)将为1GiB/秒。在分条配置中部署时,I/O子系统将并行化跨磁盘的写入,有效地将单个磁盘的吞吐量乘以配置中的条带数。因此,同样的,40MB/s磁盘将允许有效吞吐量120MB/s的RAID-0或RAID-1+0配置(带三个条带),或200MB/s(带五个条带);使用固态硬盘和/或NVMe,您可以轻松地达到1GiB/s。" #. type: Plain text #: throughput.adoc:33 msgid "" "A RAID-controller with RAM and a <> can speed up short " "spikes (by buffering them), and so too-short benchmark tests might show " "speeds like 1GiB/s too; for sustained writes its buffers will just run full," " and then not be of much help, though." msgstr "" "一个带有RAM和<>的RAID控制器可以加速短峰值(通过缓冲它们),因此太短的基准测试可能也会显示类似于1GiB/s的速度;对于持续的写操作,它的缓冲区只会满负荷运行,然后就没有多大帮助。" #. type: Plain text #: throughput.adoc:39 msgid "" "Disk _mirroring_ (RAID-1) in hardware typically has little, if any, effect " "on throughput. Disk _striping with parity_ (RAID-5) does have an effect on " "throughput, usually an adverse one when compared to striping; RAID-5 and " "RAID-6 in software even more so." msgstr "" "硬件中的磁盘镜像(RAID-1)通常对吞吐量的影响很小(如果有的话)。带奇偶校验的磁盘条带(RAID-5)确实对吞吐量有影响,与条带相比通常是不利的;软件中的RAID-5和RAID-6更是如此。" #. type: Block title #: throughput.adoc:40 #, no-wrap msgid "Network throughput" msgstr "网络吞吐量" #. type: Plain text #: throughput.adoc:50 msgid "" "indexterm:[throughput]Network throughput is usually determined by the amount" " of traffic present on the network, and on the throughput of any " "routing/switching infrastructure present. These concerns are, however, " "largely irrelevant in DRBD replication links which are normally dedicated, " "back-to-back network connections. Thus, network throughput may be improved " "either by switching to a higher-throughput hardware (such as 10 Gigabit " "Ethernet, or 56GiB Infiniband), or by using link aggregation over several " "network links, as one may do using the Linux indexterm:[bonding " "driver]`bonding` network driver." msgstr "" "indexterm:[throughput]网络吞吐量通常由网络上存在的流量以及存在的任何路由/交换基础设施的吞吐量决定。然而,这些问题在通常是专用的back-to-back网络连接的DRBD复制链路中基本上是不相关的。因此,可以通过切换到更高吞吐量的硬件(例如10 Gigabit以太网或56 GiBInfiniband)或通过使用多个网络链路上的链路聚合来提高网络吞吐量,就像可以使用Linux" " indexterm:[bonding driver]`bonding` 网络驱动程序那样。" #. type: Title === #: throughput.adoc:52 #, no-wrap msgid "Throughput overhead expectations" msgstr "吞吐量开销预期" #. type: Plain text #: throughput.adoc:56 msgid "" "When estimating the throughput overhead associated with DRBD, it is " "important to consider the following natural limitations:" msgstr "在估计与DRBD相关联的吞吐量开销时,必须考虑以下自然限制:" #. type: Plain text #: throughput.adoc:58 msgid "DRBD throughput is limited by that of the raw I/O subsystem." msgstr "DRBD吞吐量受到原始I/O子系统的限制。" #. type: Plain text #: throughput.adoc:59 msgid "DRBD throughput is limited by the available network bandwidth." msgstr "DRBD的吞吐量受到可用网络带宽的限制。" #. type: Plain text #: throughput.adoc:64 msgid "" "The _lower_ of these two establishes the theoretical throughput _maximum_ " "available to DRBD. DRBD then reduces that throughput number by its " "additional overhead, which can be expected to be less than 3 percent." msgstr "这两个值的 _lower_ 建立了DRBD可用的理论吞吐量 _最大值_ 。然后,DRBD通过其额外的开销(可以预期小于3%)来减少吞吐量。" #. type: Plain text #: throughput.adoc:71 msgid "" "Consider the example of two cluster nodes containing I/O subsystems capable " "of 600 MB/s throughput, with a Gigabit Ethernet link available between them." " Gigabit Ethernet can be expected to produce 110 MB/s throughput for TCP " "connections, thus the network connection would be the bottleneck in this " "configuration and one would expect about 110 MB/s maximum DRBD throughput." msgstr "" "以两个集群节点为例,它们包含能够达到600 " "MB/s吞吐量的I/O子系统,并且在它们之间有一个千兆以太网链路。千兆位以太网可以预期为TCP连接产生110MB/s的吞吐量,因此网络连接将是此配置中的瓶颈,并且可以预期最大DRBD吞吐量约为110 MB/s。" #. type: Plain text #: throughput.adoc:75 msgid "" "By contrast, if the I/O subsystem is capable of only 80 MB/s for sustained " "writes, then it constitutes the bottleneck, and you should expect only about" " 77 MB/s maximum DRBD throughput." msgstr "" "相比之下,如果I/O子系统仅能够以80 MB/s的速度进行持续的写操作,那么它就构成了瓶颈,您应该只期望大约77 MB/s的最大DRBD吞吐量。" #. type: Title === #: throughput.adoc:78 #, no-wrap msgid "Tuning recommendations" msgstr "调整建议" #. type: Plain text #: throughput.adoc:88 msgid "" "DRBD offers a number of configuration options which may have an effect on " "your system's throughput. This section list some recommendations for tuning " "for throughput. However, since throughput is largely hardware dependent, the" " effects of tweaking the options described here may vary greatly from system" " to system. It is important to understand that these recommendations should " "not be interpreted as \"silver bullets\" which would magically remove any " "and all throughput bottlenecks." msgstr "" "DRBD提供了许多配置选项,这些选项可能会影响系统的吞吐量。本节列出了一些有关调整吞吐量的建议。然而,由于吞吐量在很大程度上依赖于硬件,因此调整此处描述的选项的效果可能因系统而异。重要的是要明白,这些建议不应被解释为能神奇地消除任何和所有吞吐量瓶颈的 `银弹` 。" #. type: Title ==== #: throughput.adoc:90 #, no-wrap msgid "Setting `max-buffers` and `max-epoch-size`" msgstr "设置 `max-buffers` 和 `max-epoch-size` " #. type: Plain text #: throughput.adoc:100 msgid "" "These options affect write performance on the secondary nodes. `max-buffers`" " is the maximum number of buffers DRBD allocates for writing data to disk " "while `max-epoch-size` is the maximum number of write requests permitted " "between two write barriers. `max-buffers` must be equal or bigger to `max-" "epoch-size` to increase performance. The default for both is 2048; setting " "it to around 8000 should be fine for most reasonably high-performance " "hardware RAID controllers." msgstr "" "这些选项会影响辅助节点上的写入性能。`max buffers` 是DRBD为将数据写入磁盘而分配的最大缓冲区数,而 `max-epoch-" "size` 是两个写入屏障之间允许的最大写入请求数。`max buffers` 必须等于或大于 `max-epoch-" "size` ,才能提高性能。两者的默认值都是2048;对于大多数合理的高性能硬件RAID控制器来说,将其设置为8000左右应该没问题。" #. type: delimited block - #: throughput.adoc:111 #, no-wrap msgid "" "resource {\n" " net {\n" " max-buffers 8000;\n" " max-epoch-size 8000;\n" " ...\n" " }\n" " ...\n" "}\n" msgstr "" "resource {\n" " net {\n" " max-buffers 8000;\n" " max-epoch-size 8000;\n" " ...\n" " }\n" " ...\n" "}\n" #. type: Title ==== #: throughput.adoc:114 #, no-wrap msgid "Tuning the TCP send buffer size" msgstr "调整TCP发送缓冲区大小" #. type: Plain text #: throughput.adoc:123 msgid "" "The TCP send buffer is a memory buffer for outgoing TCP traffic. By default," " it is set to a size of 128 KiB. For use in high-throughput networks (such " "as dedicated Gigabit Ethernet or load-balanced bonded connections), it may " "make sense to increase this to a size of 2MiB, or perhaps even more. Send " "buffer sizes of more than 16MiB are generally not recommended (and are also " "unlikely to produce any throughput improvement)." msgstr "" "TCP发送缓冲区是用于传出TCP通信的内存缓冲区。默认情况下,它的大小设置为128 " "KiB。对于在高吞吐量网络(如专用千兆以太网或负载平衡的绑定连接)中使用,将其大小增加到2MiB或更大可能是有意义的。通常不建议发送缓冲区大小超过16MiB(而且也不太可能产生任何吞吐量改进)。" #. type: delimited block - #: throughput.adoc:133 #, no-wrap msgid "" "resource {\n" " net {\n" " sndbuf-size 2M;\n" " ...\n" " }\n" " ...\n" "}\n" msgstr "" "resource {\n" " net {\n" " sndbuf-size 2M;\n" " ...\n" " }\n" " ...\n" "}\n" #. type: Plain text #: throughput.adoc:139 msgid "" "DRBD also supports TCP send buffer auto-tuning. After enabling this feature," " DRBD will dynamically select an appropriate TCP send buffer size. TCP send " "buffer auto tuning is enabled by simply setting the buffer size to zero:" msgstr "" "DRBD还支持TCP发送缓冲区自动调整。启用此功能后,DRBD将动态选择适当的TCP发送缓冲区大小。只需将缓冲区大小设置为零即可启用TCP发送缓冲区自动调整:" #. type: delimited block - #: throughput.adoc:149 #, no-wrap msgid "" "resource {\n" " net {\n" " sndbuf-size 0;\n" " ...\n" " }\n" " ...\n" "}\n" msgstr "" "resource {\n" " net {\n" " sndbuf-size 0;\n" " ...\n" " }\n" " ...\n" "}\n" #. type: Plain text #: throughput.adoc:155 msgid "" "Please note that your ``sysctl``'s settings `net.ipv4.tcp_rmem` and " "`net.ipv4.tcp_wmem` will still influence the behaviour; you should check " "these settings, and perhaps set them similar to `131072 1048576 16777216` " "(minimum 128kiB, default 1MiB, max 16MiB)." msgstr "" "请注意,您的 `sysctl` 设置 `net.ipv4.tcp_rmem` 和 `net.ipv4.tcp_wmem` 仍将影响行为;您应该检查这些设置,并可能将它们设置为类似于 `131072" " 1048576 16777216` (最小128kiB,默认1MiB,最大16MiB)。" #. type: Plain text #: throughput.adoc:159 msgid "" "`net.ipv4.tcp_mem` is a different beast, with a different unit - do not " "touch, wrong values can easily push your machine into out-of-memory " "situations!" msgstr "`net.ipv4.tcp_mem` 是另一个怪兽,有不同的单位-不要碰,错误的值很容易把你的机器推入内存不足的情况!" #. type: Title ==== #: throughput.adoc:161 #, no-wrap msgid "Tuning the Activity Log size" msgstr "调整活动日志大小" #. type: Plain text #: throughput.adoc:167 msgid "" "If the application using DRBD is write intensive in the sense that it " "frequently issues small writes scattered across the device, it is usually " "advisable to use a fairly large activity log. Otherwise, frequent metadata " "updates may be detrimental to write performance." msgstr "" "如果使用DRBD的应用程序是写密集型的,因为它经常在设备上分散地发出小的写操作,那么通常建议使用相当大的活动日志。否则,频繁的元数据更新可能会损害写入性能。" #. type: delimited block - #: throughput.adoc:177 #, no-wrap msgid "" "resource {\n" " disk {\n" " al-extents 6007;\n" " ...\n" " }\n" " ...\n" "}\n" msgstr "" "resource {\n" " disk {\n" " al-extents 6007;\n" " ...\n" " }\n" " ...\n" "}\n" #. For tuning on striped RAID setups, please see the <> #. and #. <> settings, too. #. type: Title ==== #: throughput.adoc:185 #, no-wrap msgid "Disabling barriers and disk flushes" msgstr "禁用屏障和磁盘刷新" #. type: Plain text #: throughput.adoc:189 msgid "" "The recommendations outlined in this section should be applied _only_ to " "systems with non-volatile (battery backed) controller caches." msgstr "本节概述的建议应仅适用于具有非易失性(电池支持)控制器缓存的系统。" #. type: Plain text #: throughput.adoc:194 msgid "" "Systems equipped with battery backed write cache come with built-in means of" " protecting data in the face of power failure. In that case, it is " "permissible to disable some of DRBD's own safeguards created for the same " "purpose. This may be beneficial in terms of throughput:" msgstr "" "配备电池支持的写缓存的系统配备了内置的方法,可以在断电时保护数据。在这种情况下,允许禁用为相同目的而创建的一些DRBD自己的安全措施。这可能对吞吐量有利:" #. type: delimited block - #: throughput.adoc:205 #, no-wrap msgid "" "resource {\n" " disk {\n" " disk-barrier no;\n" " disk-flushes no;\n" " ...\n" " }\n" " ...\n" "}\n" msgstr "" "resource {\n" " disk {\n" " disk-barrier no;\n" " disk-flushes no;\n" " ...\n" " }\n" " ...\n" "}\n" #. type: Title === #: throughput.adoc:209 #, no-wrap msgid "Achieving better Read Performance via increased Redundancy" msgstr "通过增加冗余实现更好的读取性能" #. type: Plain text #: throughput.adoc:213 msgid "" "As detailed in the man page of `drbd.conf` under `read-balancing`, you can " "increase your read performance by adding more copies of your data." msgstr "如 `read-balancing` 下 `drbd.conf` 的手册页所述,可以通过添加更多数据副本来提高读取性能。" #. type: Plain text #: throughput.adoc:217 msgid "" "As a ballpark figure: with a single node processing read requests, `fio` on " "a __FusionIO__ card gave us 100k IOPs; after enabling `read-balancing`, the " "performance jumped to 180k IOPs, i.e. +80%!" msgstr "" "一个大概的数字是:单节点处理读请求时,FusionIO卡上的 `fio`给了我们10万IOPs;启用 `读平衡` 后,性能跃升到18万IOPs,即+80%!" #. type: Plain text #: throughput.adoc:220 msgid "" "So, in case you're running a read-mostly workload (big databases with lots " "of random reads), it might be worth a try to turn `read-balancing` on - and," " perhaps, add another copy for still more read IO throughput." msgstr "" "因此,如果您运行的是以读为主的工作负载(具有大量随机读操作的大型数据库),那么可能值得尝试启用 `读平衡` ,并且,也许可以添加另一个副本以获得更大的读IO吞吐量。"