FIO（flexible I/O tester）是一个测试硬盘性能非常好的工具，这里测试下。

安装 FIO

安装相关包

1	yum install libaio libaio-devel

解压安装

[root@csv-tfcs01 fio-2.1.10]# cd /opt/soft_bak/fio-2.1.10
[root@csv-tfcs01 fio-2.1.10]# ./configure
[root@csv-tfcs01 fio-2.1.10]# make
[root@csv-tfcs01 fio-2.1.10]# make install

备注：FIO 的安装非常简单，接下对计划在物理机上做个 FIO 测试，物理机上用了两块900G SAS 盘做了系统盘，SSD 盘做了数据盘，计划对系统盘和数据盘做性能测试，主要测试指标：测试8K随机写、8K随机读、8K混合读写、1MB顺序写、1MB顺序读、1MB顺序读写；测试命令模板如下

FIO 测试命令

--8k  随机写
fio -name=8krandw  -runtime=120  -filename=/data/rand.txt -ioengine=libaio -direct=1  -bs=8K  -size=10g  -iodepth=128  -numjobs=1  -rw=randwrite -group_reporting -time_based
  
--8K  随机读
fio -name=8krandr  -runtime=120  -filename=/data/rand.txt -ioengine=libaio -direct=1  -bs=8K  -size=10g  -iodepth=128  -numjobs=1  -rw=randread -group_reporting -time_based
  
--8k  混合读写
fio -name=8krandrw  -runtime=120  -filename=/data/rand.txt -ioengine=libaio -direct=1  -bs=8k  -size=10g  -iodepth=128  -numjobs=1  -rw=randrw -rwmixwrite=30  -group_reporting -time_based
   
--1Mb  顺序写
fio -name=1mseqw  -runtime=120  -filename=/data/seq.txt -ioengine=libaio -direct=1  -bs=1024k  -size=20g  -iodepth=128  -numjobs=1  -rw=write -group_reporting -time_based
  
--1Mb  顺序读
fio -name=1mseqr  -runtime=120  -filename=/data/seq.txt -ioengine=libaio -direct=1  -bs=1024k  -size=20g  -iodepth=128  -numjobs=1  -rw=read -group_reporting -time_based
  
--1Mb  顺序读写
fio -name=1mseqrw  -runtime=120  -filename=/data/seq.txt -ioengine=libaio -direct=1  -bs=1024k  -size=20g  -iodepth=128  -numjobs=1  -rw=rw -rwmixwrite=30  -group_reporting -time_based

8K随机写命令输出

测试结果

设备型号：RH5885 V3；硬盘：900G/SAS 10K/2.5寸硬盘/RAID1

8k 随机写:iops=798
8k  随机读:iops=11190
8k  混合读写(读写7:3):读iops=1114  写iops=479
1m  顺序写:iops=170
1m  顺序读:iops=246
1m  顺序读写(读写7:3):读iops=134  写iops=56

型号：RH5885 V3；硬盘：480G/Intel SATA接口SSD/2.5寸硬盘/RAID5

8k 随机写:iops=12873
8k  随机读:iops=68636
8k  混合读写(读写7:3):读iops=24846  写iops=10643
1m  顺序写:iops=1687
1m  顺序读:iops=2347
1m  顺序读写(读写7:3):读iops=1019  写iops=431

虚拟机 4c/8G 接软件定义存储

8k  随机写:iops=20234  
8k  随机读:iops=20192  
8k  混合读写:读iops=14510  写iops=6215  
1m  顺序写:iops=923  
1m  顺序读:iops=699  
1m  顺序读写:读iops=571  写iops=241

备注：从测试结果来看， SATA 接口的 SSD 盘 8K随机读写、1M顺序读写能力远远超过 SAS 盘

附：FIO 相关参数

readwrite=str, rw=str
Type of I/O pattern. Accepted values are:
  
read Sequential reads.
  
        write Sequential writes.
  
trim Sequential trim (Linux block devices only).
  
        randread
Random reads.
        randwrite
Random writes.
        randtrim
Random trim (Linux block devices only).
        rw, readwrite
 Mixed sequential reads and writes.
  
randrw Mixed random reads and writes.
  
direct=bool
 If  true,  use non-buffered I/O (usually O_DIRECT). Default:  false.
iodepth=int
 Number of I/O units to keep in flight against the file.  Note that increasing iodepth beyond 1 willnot
affect synchronous ioengines (except  for small degress when verify_async is  in  use).  Even async engines
may impose OS restrictions causing the desired depth not to be achieved. This may happen on Linux when
 using libaio and  not setting direct=1, since buffered IO is  not async on that OS.  Keep an eye on the IO
depth distribution in the fio output to verify that the achieved depth is  as expected.  Default:  1. 
ioengine=str
 Defines how the job issues I/O. The following types are defined:
sync Basic read(2)  or write(2) I/O. fseek(2)  is used to position the I/O location.
           psync Basic pread(2)  or pwrite(2) I/O.
           vsync Basic readv(2) or writev(2) I/O.  Will emulate queuing by coalescing adjacent IOs  into a
              single submission.
pvsync Basic preadv(2)  or pwritev(2) I/O.
libaio Linux  native asynchronous I/O.  This ioengine defines engine specific options.
blocksize=int[,int], bs=int[,int]
 Block size for I/O units. Default:  4k. Values  for reads, writes,  and trims can be specified separately
 in the format read,write,trim either of which may be empty to leave that value at its default.  If a
       trailing comma isn’t given, the remainder will inherit the last value set.
size=int
 Total size of I/O for  this job. fio will run until  this many bytes have been transferred, unless lim-
       ited by other options(runtime, for instance). Unless nrfiles and filesize options are given,  this
amount will be divided between the available files for the job.  If  not  set, fio will use the full size
       of the given files or devices.  If the files do  not exist, size must be given.  It  is also possible to
give size as a percentage between 1  and  100.  If size=20%  is given, fio will use  20% of the full size of
the given files or devices.
numjobs=int
 Number of clones (processes/threads performing the same workload) of this job. Default:  1.

FIO之一：硬盘性能测试工具FIO初步使用

安装 FIO

FIO 测试命令

8K随机写命令输出

测试结果

附：FIO 相关参数

参考