13.19. File system test

写写空文件

$ dd bs=1 seek=2TB if=/dev/null of=test
$ time dd if=/dev/zero of=/srv/file bs=1M count=1000
		

写随机文件

$ time dd if=/dev/urandom of=test.txt bs=1M count=1000
		

13.19.1. ext4 vs btrfs

$ cat /etc/fstab
# /etc/fstab: static file system information.
#
# Use 'blkid -o value -s UUID' to print the universally unique identifier
# for a device; this may be used with UUID= as a more robust way to name
# devices that works even if disks are added and removed. See fstab(5).
#
# <file system> <mount point>   <type>  <options>       <dump>  <pass>
proc            /proc           proc    nodev,noexec,nosuid 0       0
/dev/sda1       /               ext4    errors=remount-ro 0       1
# /opt was on /dev/sda7 during installation
UUID=5ce518a4-0f46-4688-8002-84bac2330282 /opt            btrfs   defaults        0       2
# /srv was on /dev/sda6 during installation
UUID=19573a64-f0a6-4250-a9fd-532e3d4e3477 /srv            ext4   defaults        0       2
# swap was on /dev/sda5 during installation
UUID=0f2e2f50-d989-47bf-afb7-7593888222cf none            swap    sw              0       0
neo@neo-Vostro-3400:~$ time dd if=/dev/zero of=/srv/file bs=1M count=100
100+0 records in
100+0 records out
104857600 bytes (105 MB) copied, 0.500941 s, 209 MB/s

real	0m0.521s
user	0m0.000s
sys	0m0.140s
neo@neo-Vostro-3400:~$ time dd if=/dev/zero of=/srv/file bs=1M count=100
100+0 records in
100+0 records out
104857600 bytes (105 MB) copied, 0.672553 s, 156 MB/s

real	0m0.698s
user	0m0.000s
sys	0m0.160s
neo@neo-Vostro-3400:~$ time dd if=/dev/zero of=/opt/file bs=1M count=100
100+0 records in
100+0 records out
104857600 bytes (105 MB) copied, 0.0987276 s, 1.1 GB/s

real	0m0.133s
user	0m0.000s
sys	0m0.120s
neo@neo-Vostro-3400:~$ time dd if=/dev/zero of=/opt/file bs=1M count=100
100+0 records in
100+0 records out
104857600 bytes (105 MB) copied, 0.101664 s, 1.0 GB/s

real	0m0.134s
user	0m0.000s
sys	0m0.140s

neo@neo-Vostro-3400:~$ time dd if=/dev/zero of=/srv/file bs=1M count=1000
1000+0 records in
1000+0 records out
1048576000 bytes (1.0 GB) copied, 11.8609 s, 88.4 MB/s

real	0m11.914s
user	0m0.010s
sys	0m1.360s
neo@neo-Vostro-3400:~$ time dd if=/dev/zero of=/opt/file bs=1M count=1000
1000+0 records in
1000+0 records out
1048576000 bytes (1.0 GB) copied, 9.80331 s, 107 MB/s

real	0m9.860s
user	0m0.000s
sys	0m0.880s
neo@neo-Vostro-3400:~$

13.19.2. xfs vs jfs vs reiserfs

$ cat /etc/fstab
# /etc/fstab: static file system information.
#
# <file system> <mount point>   <type>  <options>       <dump>  <pass>
proc            /proc           proc    defaults        0       0
# /dev/sda2       /               ext3    errors=remount-ro 0       1
UUID=8ce10c79-f97e-4585-8c07-75f64f043137       /               ext3    errors=remount-ro 0       1
# /dev/sda1       /boot           ext3    defaults        0       2
UUID=35705945-65ed-437b-9f79-fd0e014d100c       /boot           ext3    defaults        0       2
# /dev/sda5       /home           reiserfs defaults        0       2
UUID=f376adb7-e943-4805-892a-4fa457150b66       /home           reiserfs defaults        0       2
# /dev/sda7       /srv            xfs     defaults        0       2
UUID=2ee5c516-707f-47dc-a6a6-0d49d5dc9829       /srv            xfs     defaults        0       2
# /dev/sda8       /var            jfs     defaults        0       2
UUID=2928ba86-72fb-4b60-adc2-0f7d47e62d03       /var            jfs     defaults        0       2
# /dev/sda6       /var/www        ext3    defaults        0       2
UUID=34d3890f-a682-42ea-bdca-815f442e6539       /var/www        ext3    defaults        0       2
# /dev/sda3       none            swap    sw              0       0
UUID=bf605f47-70bc-4653-be98-c8659f959e25       none            swap    sw              0       0
/dev/scd0       /media/cdrom0   udf,iso9660 user,noauto     0       0
# XFS

neo@deployment:~$ time dd if=/dev/zero of=/srv/file bs=1M count=100
100+0 records in
100+0 records out
104857600 bytes (105 MB) copied, 0.0897329 s, 1.2 GB/s

real	0m0.117s
user	0m0.000s
sys	0m0.088s
neo@deployment:~$ time dd if=/dev/zero of=/srv/file bs=1M count=1000
1000+0 records in
1000+0 records out
1048576000 bytes (1.0 GB) copied, 4.86382 s, 216 MB/s

real	0m4.885s
user	0m0.000s
sys	0m0.960s

# JFS

neo@deployment:~$ time dd if=/dev/zero of=/var/tmp/file bs=1M count=100
100+0 records in
100+0 records out
104857600 bytes (105 MB) copied, 0.127605 s, 822 MB/s

real	0m0.157s
user	0m0.000s
sys	0m0.100s
neo@deployment:~$ time dd if=/dev/zero of=/var/tmp/file bs=1M count=1000
1000+0 records in
1000+0 records out
1048576000 bytes (1.0 GB) copied, 9.58573 s, 109 MB/s

real	0m9.597s
user	0m0.000s
sys	0m0.988s

# reiserfs

neo@deployment:~$ time dd if=/dev/zero of=/home/neo/file bs=1M count=100
100+0 records in
100+0 records out
104857600 bytes (105 MB) copied, 0.392038 s, 267 MB/s

real	0m0.430s
user	0m0.000s
sys	0m0.252s
neo@deployment:~$ time dd if=/dev/zero of=/home/neo/file bs=1M count=1000
1000+0 records in
1000+0 records out
1048576000 bytes (1.0 GB) copied, 4.62378 s, 227 MB/s

real	0m4.663s
user	0m0.000s
sys	0m2.592s

# EXT3

neo@deployment:~$ time dd if=/dev/zero of=/var/www/file bs=1M count=100
100+0 records in
100+0 records out
104857600 bytes (105 MB) copied, 0.189314 s, 554 MB/s

real	0m0.207s
user	0m0.004s
sys	0m0.176s
neo@deployment:~$ time dd if=/dev/zero of=/var/www/file bs=1M count=1000
1000+0 records in
1000+0 records out
1048576000 bytes (1.0 GB) copied, 6.46036 s, 162 MB/s

real	0m7.460s
user	0m0.008s
sys	0m1.832s

13.19.3. RAID10 (146G*8) vs EMC VNX 5300 (8G Fibre Channel)

服务器RAID卡带宽是6G,而Fibre Channel目前是8G,ISCSI与FCoE 可以提供 10G带宽,InfiniBand可以提供120G带宽。

# cat /etc/fstab
LABEL=/                 /                       ext3    defaults        1 1
LABEL=/boot             /boot                   ext3    defaults        1 2
tmpfs                   /dev/shm                tmpfs   defaults        0 0
devpts                  /dev/pts                devpts  gid=5,mode=620  0 0
sysfs                   /sys                    sysfs   defaults        0 0
proc                    /proc                   proc    defaults        0 0
LABEL=SWAP-sda3         swap                    swap    defaults        0 0
/dev/sda4               /home/oracle/rman       ext3    defaults        0 2
/dev/sdb1               /opt/oracle     ext3    defaults        0 2
/dev/emcpowerj1         /u02                    ext3    defaults        0 0

# df -h
Filesystem            Size  Used Avail Use% Mounted on
/dev/sda2              95G  7.8G   82G   9% /
/dev/sda1             1.9G   42M  1.8G   3% /boot
tmpfs                  63G  534M   63G   1% /dev/shm
/dev/sda4             924G  619G  258G  71% /home/oracle/rman
/dev/sdb1             1.1T  309G  735G  30% /opt/oracle
/dev/emcpowerj1       296G   20G  261G   7% /u02
			

IBM X3850 G5

# time dd if=/dev/zero of=file bs=1M count=100
100+0 records in
100+0 records out
104857600 bytes (105 MB) copied, 0.152702 seconds, 687 MB/s

real	0m0.154s
user	0m0.001s
sys	0m0.153s
# time dd if=/dev/zero of=file bs=1M count=1000
1000+0 records in
1000+0 records out
1048576000 bytes (1.0 GB) copied, 1.6589 seconds, 632 MB/s

real	0m1.710s
user	0m0.009s
sys	0m1.657s

# time dd if=/dev/zero of=file bs=1G count=2
2+0 records in
2+0 records out
2147483648 bytes (2.1 GB) copied, 3.48809 seconds, 616 MB/s

real	0m5.899s
user	0m0.000s
sys	0m5.594s
			

EMC

# time dd if=/dev/zero of=file bs=1M count=100
100+0 records in
100+0 records out
104857600 bytes (105 MB) copied, 0.175535 seconds, 597 MB/s

real	0m0.178s
user	0m0.001s
sys	0m0.171s
# time dd if=/dev/zero of=file bs=1M count=1000
1000+0 records in
1000+0 records out
1048576000 bytes (1.0 GB) copied, 1.67429 seconds, 626 MB/s

real	0m1.718s
user	0m0.002s
sys	0m1.664s
# time dd if=/dev/zero of=file bs=1G count=2
2+0 records in
2+0 records out
2147483648 bytes (2.1 GB) copied, 3.46919 seconds, 619 MB/s

real	0m3.757s
user	0m0.002s
sys	0m3.656s
			

13.19.4. Dell 2950(RAID5 500G SATA * 6) vs MD1200

# cat /etc/fstab
# /etc/fstab: static file system information.
#
# Use 'blkid -o value -s UUID' to print the universally unique identifier
# for a device; this may be used with UUID= as a more robust way to name
# devices that works even if disks are added and removed. See fstab(5).
#
# <file system> <mount point>   <type>  <options>       <dump>  <pass>
proc            /proc           proc    nodev,noexec,nosuid 0       0
# / was on /dev/sda1 during installation
UUID=2fc411ec-9f6e-4e04-9270-11d23a9b0668 /               ext4    errors=remount-ro 0       1
# swap was on /dev/sda2 during installation
UUID=f5175b7a-4c87-471c-ab9f-9d601bc5e6e2 none            swap    sw              0       0
UUID=3217bdd9-1beb-494a-a428-8d1c09eaa1af /backup ext4 errors=remount-ro 0       1
UUID=9bed3b85-bbc5-4aec-8c9a-8911712ea0c6 /backup1 ext4 errors=remount-ro 0       1
UUID=24bec385-2074-4eb5-8d24-22c33dc245d8 /backup2 ext4 errors=remount-ro 0       1

# df -h
Filesystem            Size  Used Avail Use% Mounted on
/dev/sda1              46G  8.6G   35G  20% /
none                  2.0G  212K  2.0G   1% /dev
none                  2.0G     0  2.0G   0% /dev/shm
none                  2.0G  9.0M  2.0G   1% /var/run
none                  2.0G     0  2.0G   0% /var/lock
none                   46G  8.6G   35G  20% /var/lib/ureadahead/debugfs
/dev/sda3             2.2T  2.0T   89G  96% /backup
/dev/sdc1             7.2T  6.0T  887G  88% /backup2
/dev/sdb1             9.0T  7.0T  1.6T  83% /backup1

/dev/sda 是2950 RAID5 500G*6 1000RPM

/dev/sdb 是MD1200 RAID5 2T*6 7200RPM

/dev/sdc 是MD1200 RAID50 2T*6 7200RPM

root@backup:~# time dd if=/dev/zero of=/backup/file bs=1M count=100
100+0 records in
100+0 records out
104857600 bytes (105 MB) copied, 0.156596 s, 670 MB/s

real	0m0.242s
user	0m0.010s
sys	0m0.150s
root@backup:~# time dd if=/dev/zero of=/backup/file bs=1M count=1000
1000+0 records in
1000+0 records out
1048576000 bytes (1.0 GB) copied, 4.61282 s, 227 MB/s

real	0m4.763s
user	0m0.000s
sys	0m1.640s
root@backup:~# time dd if=/dev/zero of=/backup/file bs=1G count=5
5+0 records in
5+0 records out
5368709120 bytes (5.4 GB) copied, 33.7263 s, 159 MB/s

real	0m34.685s
user	0m0.000s
sys	0m13.070s
root@backup:~# time dd if=/dev/zero of=/backup1/file bs=1M count=100
100+0 records in
100+0 records out
104857600 bytes (105 MB) copied, 0.130451 s, 804 MB/s

real	0m0.290s
user	0m0.000s
sys	0m0.130s
root@backup:~# time dd if=/dev/zero of=/backup1/file bs=1M count=1000
1000+0 records in
1000+0 records out
1048576000 bytes (1.0 GB) copied, 57.1654 s, 18.3 MB/s

real	0m57.206s
user	0m0.000s
sys	0m1.580s
root@backup:~# time dd if=/dev/zero of=/backup1/file bs=1G count=5
5+0 records in
5+0 records out
5368709120 bytes (5.4 GB) copied, 309.194 s, 17.4 MB/s

real	5m9.762s
user	0m0.000s
sys	0m10.820s
root@backup:~# time dd if=/dev/zero of=/backup2/file bs=1M count=100
100+0 records in
100+0 records out
104857600 bytes (105 MB) copied, 0.145224 s, 722 MB/s

real	0m0.333s
user	0m0.000s
sys	0m0.130s
root@backup:~# time dd if=/dev/zero of=/backup2/file bs=1M count=1000
1000+0 records in
1000+0 records out
1048576000 bytes (1.0 GB) copied, 41.9185 s, 25.0 MB/s

real	0m41.979s
user	0m0.010s
sys	0m1.930s
root@backup:~# time dd if=/dev/zero of=/backup2/file bs=1G count=5
5+0 records in
5+0 records out
5368709120 bytes (5.4 GB) copied, 200.384 s, 26.8 MB/s

real	3m20.850s
user	0m0.000s
sys	0m12.490s

原文出处:Netkiller 系列 手札
本文作者:陈景峯
转载请与作者联系,同时请务必标明文章原始出处和作者信息及本声明。

时间: 2024-11-09 05:10:50

13.19. File system test的相关文章

第 13 章 File System

13.1. /etc/fstab # <file system> <mount point> <type> <options> <dump> <pass> mount point 该字段描述希望的文件系统加载的目录,对于swap设备,该字段为none file system 例如/dev/cdrom或/dev/sdb,除了使用设备名,你可以使用设备的UUID或设备的卷标签,例如,LABAL=root 或 UUID=7f91104e-8

Moose File System

Moose File System http://www.moosefs.org/ 4.1. Master server installation groupadd mfs useradd -g mfs mfs cd /usr/local/src wget http://pro.hit.gemius.pl/hitredir/id=nXCV9nrckU2Et.zoR5kxdXZJLQq1fqbG4AIiq5K95Gz.07/url=moosefs.org/tl_files/mfscode/mfs-

27.3. Directory and File System Related

27.3.1. dirname $ dirname /usr/bin/find /usr/bin 27.3.2. filename $ basename /usr/bin/find find 27.3.2.1. 排除扩展名 file=test.txt b=${file%.*} echo $b $ for file in *.JPG;do mv $file ${file%.*}.jpg;done 27.3.2.2. 取扩展名 file=test.txt b=${file##*.} echo $b

5.3. Directory and File System Related

5.3.1. dirname $ dirname /usr/bin/find /usr/bin 5.3.2. filename $ basename /usr/bin/find find 5.3.2.1. 排除扩展名 file=test.txt b=${file%.*} echo $b $ for file in *.JPG;do mv $file ${file%.*}.jpg;done 5.3.2.2. 取扩展名 file=test.txt b=${file##*.} echo $b 5.3.

nilfs (a continuent snapshot file system) used with PostgreSQL

NILFS 是NTT开发的一个文件系统, 持续的提供快照, 为误操作提供回滚可能.  相比lvm的快照, nilfs更有利于SSD环境. 因为在写频繁的环境中不需要大量的cow.  简介 :  http://nilfs.sourceforge.net/en/ NILFS is a log-structured file system supporting versioning of the entire file system and continuous snapshotting which

利用 mount 指令解决 Read-only file system的问题

在linux系统中创建一个文件提示: /application/report/shiwei # touch test.c touch: cannot touch `test.c': Read-only file system   总是提示Read-only file system,也就是说系统是只读的,什么也写不了.于是在网上到处找了一下,发现解决起来挺容易的.一条命令就可以了   mount -o remount rw / 参考:http://www.blogjava.net/hadeslee

android开发-安卓开发中read-only file system

问题描述 安卓开发中read-only file system android模拟机上不能加文件提示read only file system 是什么问题啊 mount -o remount ,rw /都试过了 ,没什么用额. 使用重新挂载还是无用 mount -o remount,rw rootfs /system/ mount -o remount,rw rootfs /system/ cd sdcard cd sdcard mkdir a mkdir a mkdir failed for

asm和file system之间数据文件的转换

How to move a datafile from a file system to ASM Moving a datafile from the file system can be achived in two ways. i. While the database is shutdown (in mount stage). ii. While the database is running (with the selected tablespace offline). (数据文件实现文

linux系统启动报错:[contains a file system with errors, check forced]的解决方法

  linux系统启动报错:[contains a file system with errors, check forced]的解决方法参考 1.解决参考一Press enter for maintenance(or type Control-D to continue):/dev/sda3 contains a file system with errors, check forced./dev/sda3: Inodes that were part of a corrupted orpha