update kernel tuning zh_CN document (#10433)
飞雪无情
4 years ago
committed by
GitHub
parent
9f60e84ce1
commit
a694ba93d9
@ -1,71 +1,96 @@ |
||||
# Linux服务器上MinIO生产环境的内核调优 [](https://slack.min.io) [](https://hub.docker.com/r/minio/minio/) |
||||
|
||||
## 调优网络参数 |
||||
这儿有一份针对MinIO服务器内核调优的建议， 你可以拷贝这个[脚本](https://github.com/minio/minio/blob/master/docs/deployment/kernel-tuning/sysctl.sh)到你的服务器上使用。 |
||||
|
||||
以下网络参数设置可帮助确保Minio服务器在生产环境负载上的最佳性能。 |
||||
> 注意: 这是Linux服务器上的通用建议，不过在使用前也要非常小心。这些设置不是强制性的，而且也不能解决硬件的问题，所以不要使用它们提高 |
||||
> 性能掩盖硬件本身的问题。在任何情况下，都应该先进行硬件基准测试，达到预期结果后才真正的执行优化。 |
||||
|
||||
- *`tcp_fin_timeout`* : 一个socket连接大约需要1.5KB的内存，关闭未使用的socket连接可以减少内存占用，避免出现内存泄露。即使另一方由于某种原因没有关闭socket连接，系统本身也会在到达超时时间时断开连接。 `tcp_fin_timeout`参数定义了内核保持sockets在FIN-WAIT-2状态的超时时间。我们建议设成20，你可以按下面的示例进行设置。 |
||||
|
||||
```sh |
||||
sysctl -w net.ipv4.tcp_fin_timeout=30 |
||||
``` |
||||
#!/bin/bash |
||||
|
||||
- *`tcp_keepalive_probes`* : 这个参数定义了经过几次无回应的探测之后，认为连接断开了。你可以按下面的示例进行设置。 |
||||
cat > sysctl.conf <<EOF |
||||
# maximum number of open files/file descriptors |
||||
fs.file-max = 4194303 |
||||
|
||||
```sh |
||||
sysctl -w net.ipv4.tcp_keepalive_probes=5 |
||||
``` |
||||
# use as little swap space as possible |
||||
vm.swappiness = 1 |
||||
|
||||
- *`wmem_max`*: 这个参数定义了针对所有类型的连接，操作系统的最大发送buffer大小。 |
||||
# prioritize application RAM against disk/swap cache |
||||
vm.vfs_cache_pressure = 50 |
||||
|
||||
```sh |
||||
sysctl -w net.core.wmem_max=540000 |
||||
``` |
||||
# minimum free memory |
||||
vm.min_free_kbytes = 1000000 |
||||
|
||||
- *`rmem_max`*: 这个参数定义了针对所有类型的连接，操作系统最大接收buffer大小。 |
||||
# follow mellanox best practices https://community.mellanox.com/s/article/linux-sysctl-tuning |
||||
# the following changes are recommended for improving IPv4 traffic performance by Mellanox |
||||
|
||||
```sh |
||||
sysctl -w net.core.rmem_max=540000 |
||||
``` |
||||
# disable the TCP timestamps option for better CPU utilization |
||||
net.ipv4.tcp_timestamps = 0 |
||||
|
||||
## 调优虚拟内存 |
||||
# enable the TCP selective acks option for better throughput |
||||
net.ipv4.tcp_sack = 1 |
||||
|
||||
下面是推荐的虚拟内存设置。 |
||||
# increase the maximum length of processor input queues |
||||
net.core.netdev_max_backlog = 250000 |
||||
|
||||
- *`swappiness`* : 此参数控制了交换运行时内存的相对权重，而不是从page缓存中删除page。取值范围是[0,100],我们建议设成10。 |
||||
# increase the TCP maximum and default buffer sizes using setsockopt() |
||||
net.core.rmem_max = 4194304 |
||||
net.core.wmem_max = 4194304 |
||||
net.core.rmem_default = 4194304 |
||||
net.core.wmem_default = 4194304 |
||||
net.core.optmem_max = 4194304 |
||||
|
||||
```sh |
||||
sysctl -w vm.swappiness=10 |
||||
``` |
||||
# increase memory thresholds to prevent packet dropping: |
||||
net.ipv4.tcp_rmem = "4096 87380 4194304" |
||||
net.ipv4.tcp_wmem = "4096 65536 4194304" |
||||
|
||||
- *`dirty_background_ratio`*: 这个是`脏`页可以占系统内存的百分比，内存页仍需要写到磁盘上。我们建议要尽早将数据写到磁盘上，越早越好。为了达到这个目的，将 `dirty_background_ratio` 设成1。 |
||||
# enable low latency mode for TCP: |
||||
net.ipv4.tcp_low_latency = 1 |
||||
|
||||
```sh |
||||
sysctl -w vm.dirty_background_ratio=1 |
||||
``` |
||||
# the following variable is used to tell the kernel how much of the socket buffer |
||||
# space should be used for TCP window size, and how much to save for an application |
||||
# buffer. A value of 1 means the socket buffer will be divided evenly between. |
||||
# TCP windows size and application. |
||||
net.ipv4.tcp_adv_win_scale = 1 |
||||
|
||||
- *`dirty_ratio`*: 这定义了在所有事务必须提交到磁盘之前，可以用脏页填充的系统内存的绝对最大数量。 |
||||
# maximum number of incoming connections |
||||
net.core.somaxconn = 65535 |
||||
|
||||
```sh |
||||
sysctl -w vm.dirty_ratio=5 |
||||
``` |
||||
# maximum number of packets queued |
||||
net.core.netdev_max_backlog = 10000 |
||||
|
||||
## 调优调度程序 |
||||
# queue length of completely established sockets waiting for accept |
||||
net.ipv4.tcp_max_syn_backlog = 4096 |
||||
|
||||
正确的调度程序配置确保Minio进程获得足够的CPU时间。 以下是推荐的调度程序设置。 |
||||
# time to wait (seconds) for FIN packet |
||||
net.ipv4.tcp_fin_timeout = 15 |
||||
|
||||
- *`sched_min_granularity_ns`*: 此参数定义了一个任务在被其它任务抢占时，可在CPU一次运行的最短时间，我们建议设成10ms。 |
||||
# disable icmp send redirects |
||||
net.ipv4.conf.all.send_redirects = 0 |
||||
|
||||
```sh |
||||
sysctl -w kernel.sched_min_granularity_ns=10000000 |
||||
``` |
||||
# disable icmp accept redirect |
||||
net.ipv4.conf.all.accept_redirects = 0 |
||||
|
||||
- *`sched_wakeup_granularity_ns`*: 降低该参数值可以减少唤醒延迟，提高吞吐量。 |
||||
# drop packets with LSR or SSR |
||||
net.ipv4.conf.all.accept_source_route = 0 |
||||
|
||||
```sh |
||||
sysctl -w kernel.sched_wakeup_granularity_ns=15000000 |
||||
``` |
||||
# MTU discovery, only enable when ICMP blackhole detected |
||||
net.ipv4.tcp_mtu_probing = 1 |
||||
|
||||
## 调优磁盘 |
||||
EOF |
||||
|
||||
我们将磁盘调优的建议整合到了注释完备的 [shell script](https://github.com/minio/minio/blob/master/docs/deployment/kernel-tuning/disk-tuning.sh)，敬请查看。 |
||||
echo "Enabling system level tuning params" |
||||
sysctl --quiet --load sysctl.conf && rm -f sysctl.conf |
||||
|
||||
# `Transparent Hugepage Support`*: This is a Linux kernel feature intended to improve |
||||
# performance by making more efficient use of processor’s memory-mapping hardware. |
||||
# But this may cause https://blogs.oracle.com/linux/performance-issues-with-transparent-huge-pages-thp |
||||
# for non-optimized applications. As most Linux distributions set it to `enabled=always` by default, |
||||
# we recommend changing this to `enabled=madvise`. This will allow applications optimized |
||||
# for transparent hugepages to obtain the performance benefits, while preventing the |
||||
# associated problems otherwise. Also, set `transparent_hugepage=madvise` on your kernel |
||||
# command line (e.g. in /etc/default/grub) to persistently set this value. |
||||
|
||||
echo "Enabling THP madvise" |
||||
echo madvise | sudo tee /sys/kernel/mm/transparent_hugepage/enabled |
||||
``` |
||||
|
||||
Loading…
Reference in new issue