UniverStor P20000 (ActiveScale P100) NFS性能测试

首先这是一个不严谨的测试,Client和Server都是在有生产压力的情况下做额外测试,仅使用了一个Client和Server(仅使用三个Server中的一个),结果仅供参考。

NFS Client:mirrors.nju.edu.cn 的生产虚拟机,负载不轻,VMXNET 3网卡,主机万兆网卡,通过下面一个系统节点的IP挂载NFS。
mount -t nfs -o nfsvers=3,wsize=1048576,rsize=1048576,proto=tcp,async,lookupcache=none,timeo=600 x.x.x.x:/yaoge123 /mnt

NFS Server: UniverStor P20000 (ActiveScale P100) 5.5.0.40,三分之一配置,三个系统节点六个存储节点,box.nju.edu.cn的后端存储。
access_type=RW,clients=x.x.x.x,sec=sys

Iozone: Performance Test of File I/O
        Version $Revision: 3.487 $
    Compiled for 64 bit mode.
    Build: linux-AMD64 

Contributors:William Norcott, Don Capps, Isom Crawford, Kirby Collins
             Al Slater, Scott Rhine, Mike Wisner, Ken Goss
             Steve Landherr, Brad Smith, Mark Kelly, Dr. Alain CYR,
             Randy Dunlap, Mark Montague, Dan Million, Gavin Brebner,
             Jean-Marc Zucconi, Jeff Blomberg, Benny Halevy, Dave Boone,
             Erik Habbinga, Kris Strecker, Walter Wong, Joshua Root,
             Fabrice Bacchella, Zhenghua Xue, Qin Li, Darren Sawyer,
             Vangel Bojaxhi, Ben England, Vikentsi Lapa,
             Alexey Skidanov, Sudhir Kumar.

Run began: Mon Aug 12 20:00:41 2019

Record Size 128 kB
File size set to 16777216 kB
Command line used: /root/iozone -i0 -i1 -r 128k -s 16g -t8
Output is in kBytes/sec
Time Resolution = 0.000001 seconds.
Processor cache size set to 1024 kBytes.
Processor cache line size set to 32 bytes.
File stride size set to 17 * record size.
Throughput test with 8 processes
Each process writes a 16777216 kByte file in 128 kByte records

Children see throughput for  8 initial writers  =  719571.23 kB/sec
Parent sees throughput for  8 initial writers   =  693136.82 kB/sec
Min throughput per process          =   89555.71 kB/sec 
Max throughput per process          =   90813.93 kB/sec
Avg throughput per process          =   89946.40 kB/sec
Min xfer                    = 16543360.00 kB

Children see throughput for  8 rewriters    =  721639.59 kB/sec
Parent sees throughput for  8 rewriters     =  702820.53 kB/sec
Min throughput per process          =   89371.36 kB/sec 
Max throughput per process          =   92429.21 kB/sec
Avg throughput per process          =   90204.95 kB/sec
Min xfer                    = 16223744.00 kB

Children see throughput for  8 readers      =  819373.91 kB/sec
Parent sees throughput for  8 readers       =  819029.33 kB/sec
Min throughput per process          =  101887.23 kB/sec 
Max throughput per process          =  102778.71 kB/sec
Avg throughput per process          =  102421.74 kB/sec
Min xfer                    = 16634368.00 kB

Children see throughput for 8 re-readers    =  825189.77 kB/sec
Parent sees throughput for 8 re-readers     =  824947.40 kB/sec
Min throughput per process          =  103080.87 kB/sec 
Max throughput per process          =  103243.56 kB/sec
Avg throughput per process          =  103148.72 kB/sec
Min xfer                    = 16757248.00 kB
Iozone: Performance Test of File I/O
        Version $Revision: 3.487 $
	Compiled for 64 bit mode.
	Build: linux-AMD64 

Contributors:William Norcott, Don Capps, Isom Crawford, Kirby Collins
             Al Slater, Scott Rhine, Mike Wisner, Ken Goss
             Steve Landherr, Brad Smith, Mark Kelly, Dr. Alain CYR,
             Randy Dunlap, Mark Montague, Dan Million, Gavin Brebner,
             Jean-Marc Zucconi, Jeff Blomberg, Benny Halevy, Dave Boone,
             Erik Habbinga, Kris Strecker, Walter Wong, Joshua Root,
             Fabrice Bacchella, Zhenghua Xue, Qin Li, Darren Sawyer,
             Vangel Bojaxhi, Ben England, Vikentsi Lapa,
             Alexey Skidanov, Sudhir Kumar.

Run began: Tue Aug 13 08:30:19 2019

Record Size 128 kB
File size set to 2097152 kB
Command line used: /root/iozone -i0 -i1 -i2 -r 128k -s 2g -t8
Output is in kBytes/sec
Time Resolution = 0.000001 seconds.
Processor cache size set to 1024 kBytes.
Processor cache line size set to 32 bytes.
File stride size set to 17 * record size.
Throughput test with 8 processes
Each process writes a 2097152 kByte file in 128 kByte records

Children see throughput for  8 initial writers 	=  828709.60 kB/sec
Parent sees throughput for  8 initial writers 	=  625811.47 kB/sec
Min throughput per process 			=   89453.18 kB/sec 
Max throughput per process 			=  113444.66 kB/sec
Avg throughput per process 			=  103588.70 kB/sec
Min xfer 					= 1662848.00 kB

Children see throughput for  8 rewriters 	=  850283.30 kB/sec
Parent sees throughput for  8 rewriters 	=  672423.37 kB/sec
Min throughput per process 			=  100863.54 kB/sec 
Max throughput per process 			=  115805.04 kB/sec
Avg throughput per process 			=  106285.41 kB/sec
Min xfer 					= 1830784.00 kB

Children see throughput for  8 readers 		=  799100.11 kB/sec
Parent sees throughput for  8 readers 		=  796371.21 kB/sec
Min throughput per process 			=   98683.31 kB/sec 
Max throughput per process 			=  103026.41 kB/sec
Avg throughput per process 			=   99887.51 kB/sec
Min xfer 					= 2011648.00 kB

Children see throughput for 8 re-readers 	=  824877.78 kB/sec
Parent sees throughput for 8 re-readers 	=  823455.08 kB/sec
Min throughput per process 			=  102881.04 kB/sec 
Max throughput per process 			=  103626.41 kB/sec
Avg throughput per process 			=  103109.72 kB/sec
Min xfer 					= 2088448.00 kB

Children see throughput for 8 random readers 	=  415811.13 kB/sec
Parent sees throughput for 8 random readers 	=  414555.11 kB/sec
Min throughput per process 			=   30050.71 kB/sec 
Max throughput per process 			=  142421.50 kB/sec
Avg throughput per process 			=   51976.39 kB/sec
Min xfer 					=  442880.00 kB

Children see throughput for 8 random writers 	=   32076.26 kB/sec
Parent sees throughput for 8 random writers 	=   15830.50 kB/sec
Min throughput per process 			=    2661.41 kB/sec 
Max throughput per process 			=    6257.78 kB/sec
Avg throughput per process 			=    4009.53 kB/sec
Min xfer 					= 1048320.00 kB

发表评论

电子邮件地址不会被公开。 必填项已用*标注

*

code

此站点使用Akismet来减少垃圾评论。了解我们如何处理您的评论数据