USB3 performance tests
#3
One think to keep mind is that my board was one of pre-production samples, and I had to remove some of the ESD diodes as the wrong ones had been installed on the USB3 connector. Therefore my USB3 performance specs may not reflect the production model performance. And I certainly don't think this is the case, because my USB3 JumpDrive test subject didn't reach anywhere the below read speeds of my desktop in comparison:
Code:
Command line used: iozone -e -I -a -s 100M -r 4k -r 16k -r 512k -r 1024k -r 16384k -i 0 -i 1 -i 2 random random bkwd record stride kB reclen write rewrite read reread read write read rewrite read fwrite frewrite fread freread 102400 4 12722 11718 24340 24358 6719 462 102400 16 11709 22004 80089 3163 2931 2020 102400 512 4956 21783 1652 110923 99929 4928 102400 1024 6191 21576 116329 116532 111205 3978 102400 16384 6704 21469 117321 118603 117775 7132

rock64 /w 2GB, running Ubuntu Xenial ayufan 0.2.5, against an ext4 formatted cheap Lexar 16GB S75 USB3 flash drive. Write performance is nothing special, but I have had real world 100MB/s read speed from it on my desktop.

Code:
Iozone: Performance Test of File I/O        Version $Revision: 3.429 $ Compiled for 64 bit mode. Build: linux  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. Run began: Sat Jul  1 04:04:59 2017 Include fsync in write timing O_DIRECT feature enabled Auto Mode File size set to 102400 kB Record Size 4 kB Record Size 16 kB Record Size 512 kB Record Size 1024 kB Record Size 16384 kB Command line used: iozone -e -I -a -s 100M -r 4k -r 16k -r 512k -r 1024k -r 16384k -i 0 -i 1 -i 2 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.                                                               random    random     bkwd    record    stride                                                   kB  reclen    write  rewrite    read    reread    read     write     read   rewrite      read   fwrite frewrite    fread  freread           102400       4     7816     9891    12541    12347     4743      361                                                                     102400      16    17479    18596    31016    31207    14804     1840                                                                     102400     512    20835    20269    39598    39880    38142     5381                                                                     102400    1024    15217    20433    39748    40078    39282    10174                                                                     102400   16384    15714    19912    40093    40364    40329    21689                                                           iozone test complete.

Lets see what difference mount options make - ext4 with nodiratime and noatime mount options set:

Code:
Iozone: Performance Test of File I/O Version $Revision: 3.429 $ Compiled for 64 bit mode. Build: linux 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. Run began: Sat Jul 1 04:45:18 2017 Include fsync in write timing O_DIRECT feature enabled Auto Mode File size set to 102400 kB Record Size 4 kB Record Size 16 kB Record Size 512 kB Record Size 1024 kB Record Size 16384 kB Command line used: iozone -e -I -a -s 100M -r 4k -r 16k -r 512k -r 1024k -r 16384k -i 0 -i 1 -i 2 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. random random bkwd record stride kB reclen write rewrite read reread read write read rewrite read fwrite frewrite fread freread 102400 4 7437 9725 11792 11556 4551 333 102400 16 17176 18255 29957 29993 14715 1819 102400 512 20811 20109 39708 39797 37948 6048 102400 1024 22432 14799 39986 40086 39193 7160 102400 16384 11263 19983 40172 40163 40166 21356 iozone test complete.

A comparision of the previous results against the lower USB2 port (ext4, same nodiratime, noatime mount settings):
Code:
Iozone: Performance Test of File I/O Version $Revision: 3.429 $ Compiled for 64 bit mode. Build: linux 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. Run began: Sat Jul 1 05:00:07 2017 Include fsync in write timing O_DIRECT feature enabled Auto Mode File size set to 102400 kB Record Size 4 kB Record Size 16 kB Record Size 512 kB Record Size 1024 kB Record Size 16384 kB Command line used: iozone -e -I -a -s 100M -r 4k -r 16k -r 512k -r 1024k -r 16384k -i 0 -i 1 -i 2 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. random random bkwd record stride kB reclen write rewrite read reread read write read rewrite read fwrite frewrite fread freread 102400 4 7233 7695 9076 9093 4308 431 102400 16 11562 16447 21296 21271 12569 1722 102400 512 18149 19494 35323 35328 33954 5090 102400 1024 21691 19539 35900 35994 35326 10153 102400 16384 14447 18766 36429 36444 36409 21688 iozone test complete.


And to show how the filesystem has a big impact on the results, here is the same test again, but with the drive formated as vfat.
Code:
Iozone: Performance Test of File I/O Version $Revision: 3.429 $ Compiled for 64 bit mode. Build: linux 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. Run began: Sat Jul 1 04:28:37 2017 Include fsync in write timing O_DIRECT feature enabled Auto Mode File size set to 102400 kB Record Size 4 kB Record Size 16 kB Record Size 512 kB Record Size 1024 kB Record Size 16384 kB Command line used: iozone -e -I -a -s 100M -r 4k -r 16k -r 512k -r 1024k -r 16384k -i 0 -i 1 -i 2 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. random random bkwd record stride kB reclen write rewrite read reread read write read rewrite read fwrite frewrite fread freread 102400 4 9991 11235 13096 13105 5529 447 102400 16 15100 17735 29883 29919 14464 1082 102400 512 10093 14090 39526 39633 37734 2981 102400 1024 17169 10783 39761 39907 38927 4239 102400 16384 16900 10989 40233 40312 40308 13258 iozone test complete.

And finally, a Seagate 4TB Expansion drive (mains power external hard drive) (actually got some high numbers this time).

Code:
Iozone: Performance Test of File I/O Version $Revision: 3.429 $ 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. Run began: Sat Jul 1 15:48:05 2017 Include fsync in write timing O_DIRECT feature enabled Auto Mode File size set to 102400 kB Record Size 4 kB Record Size 16 kB Record Size 512 kB Record Size 1024 kB Record Size 16384 kB Command line used: iozone -e -I -a -s 100M -r 4k -r 16k -r 512k -r 1024k -r 16384k -i 0 -i 1 -i 2 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. random random bkwd record stride kB reclen write rewrite read reread read write read rewrite read fwrite frewrite fread freread 102400 4 12722 11718 24340 24358 6719 462 102400 16 11709 22004 80089 3163 2931 2020 102400 512 4956 21783 1652 110923 99929 4928 102400 1024 6191 21576 116329 116532 111205 3978 102400 16384 6704 21469 117321 118603 117775 7132 iozone test complete.
  Reply


Messages In This Thread
USB3 performance tests - by pfeerick - 06-30-2017, 09:28 PM
RE: USB3 performance tests - by MarkHaysHarris777 - 06-30-2017, 10:03 PM
RE: USB3 performance tests - by pfeerick - 06-30-2017, 10:14 PM
RE: USB3 performance tests - by MarkHaysHarris777 - 06-30-2017, 11:22 PM
RE: USB3 performance tests - by xalius - 07-01-2017, 05:22 AM
RE: USB3 performance tests - by xalius - 07-01-2017, 08:13 AM
RE: USB3 performance tests - by fire219 - 07-01-2017, 03:03 PM
RE: USB3 performance tests - by xalius - 07-05-2017, 01:47 PM
RE: USB3 performance tests - by xalius - 07-11-2017, 07:18 AM
RE: USB3 performance tests - by rontant - 08-18-2017, 09:04 AM
RE: USB3 performance tests - by Gobam - 02-09-2018, 11:17 AM
RE: USB3 performance tests - by S3phi40T - 02-15-2018, 02:42 PM
RE: USB3 performance tests - by pedroz - 03-17-2018, 09:35 AM

Possibly Related Threads…
Thread Author Replies Views Last Post
  USB3.0 Issues at DietPi(Armbian) #22.02.1 t4_4t 3 5,916 08-15-2022, 06:58 PM
Last Post: TRS-80
  Forcing u-boot to detect USB3 storage jandastroy 8 20,087 04-02-2019, 10:06 PM
Last Post: t4_4t
Rainbow USB3 Detection Issues When Booting Without USB Device Attached Dargmuesli 0 3,481 01-14-2019, 11:26 AM
Last Post: Dargmuesli
  USB3 problems scalextrix 2 6,499 09-08-2018, 01:15 AM
Last Post: scalextrix
  Rock64 USB3 bugs NitroDeath 4 12,501 09-02-2018, 03:00 AM
Last Post: NitroDeath
  Arch Linux USB3 Issues? decontainerized 4 12,595 07-25-2018, 04:39 AM
Last Post: cooker
  Flaky USB3 on Bionic ayufan image hackish 7 14,573 06-15-2018, 08:32 AM
Last Post: fire219
Sad Rock64 4GB poor USB3 HDD performance KernelPanic 7 21,272 04-11-2018, 02:20 AM
Last Post: KernelPanic
  Rock64 as a firewall - usb3 gigE adopters jl_678 2 7,592 02-19-2018, 03:16 PM
Last Post: spawnflagger
  Root FS on usb3 drive eloranta 7 13,114 12-18-2017, 10:43 PM
Last Post: eloranta

Forum Jump:


Users browsing this thread: 1 Guest(s)