XFS的高性能设置

原文出处： Setting up XFS on Hardware RAID — the simple edition

硬盘分区的边界对齐

RAID有最小的分配单位stripe size，硬盘的最小分配单位Sector size,我们在RAID建立的逻辑硬盘上，分区的时候要对齐stripe

• 用fdisk取得sector size, 一般是512
• 取得RAID的stripe size,一般是 64K = 2^16 = 65536 bytes
• 计算每个stripe有几个sector组成 65536 / 512 = 128
• 用fdisk取得该分区的起始扇区号，比如27344896;. 27344896 / 128 = 213632

如果刚好整出，就是对好了，否则需要重新分区，对齐边界

建立XFS文件系统
需要指定两个重要参数

• RAID的stripe size
• 有效硬盘数

RAID 1+0: drives / 2.
RAID 5: drives – 1.

指定 ‘su’ and ‘sw’参数

# mkfs.xfs -d su=64k,sw=4 /dev/sda3

指定mount参数

，比如常见的noatime,nodiratime,nobarrier
/var/lib/mysql xfs nobarrier,noatime,nodiratime

设置IO scheduler
好的IO cheduler有两个选项：

Deadline is an active scheduler

noop simply means IO will be handled without rescheduling.

立刻生效:
echo noop > /sys/block/sda/queue/scheduler # update the scheduler in realtime

永久生效：在grub.conf中指定参数‘elevator=’:
kernel /boot/vmlinuz-2.6.18-53.el5 ro root=LABEL=/ noapic acpi=off rhgb quiet notsc elevator=noop



This is a complicated topic, and I’ve tried to temper the complexity with what will provide the most benefit. What has made most improvement for you that could be added without much complexity?