Benchmarking your synology


Recommended Posts

 

I would not want to disturb you but...

 

dd without direct option shows only cache speed - no more. In some configurations, you can get a linear speed without a cache higher than with a cache. In some you will receive hundreds of megabytes per second, with linear performance in units of megabytes. With the direct option (iflag = direct for reading, oflag = direct for writing), dd checks only the linear velocity. Which is absolutely not equal to the maximum speed (if we are talking about a raid on many disks, then a raid in several threads can give more speed than in one), not a real performance. link1 link2

 

So for correct measurement, we need to do a drop cache before testing.

Edited by hellion
Link to post
Share on other sites

Having some more free time on my hand, playing around with a "new" mainboard.

 

Asrock Q1900-ITX, 8GB DDR3 RAM, 

DS3617xs 6.1 Jun's Mod V1.02b, DSM 6.1.7-15284 Update 2.

1 single 2,5" 320GB/7200rpm drive, btrfs.

admin@ITX:/$ grep MHz /proc/cpuinfo
cpu MHz         : 1993.000
cpu MHz         : 1993.000
cpu MHz         : 1993.000
cpu MHz         : 1993.000
admin@ITX:/$ dd if=/dev/zero bs=1M count=1024 | md5sum
1073741824 bytes (1.1 GB) copied, 3.18849 s, 337 MB/s
admin@ITX:/$ sudo dd bs=1M count=256 if=/dev/zero of=/volume1/test conv=fdatasync
268435456 bytes (268 MB) copied, 2.78 s, 96.6 MB/s
1073741824 bytes (1.1 GB) copied, 11.1011 s, 96.7 MB/s
2147483648 bytes (2.1 GB) copied, 20.899 s, 103 MB/s
44294967296 bytes (4.3 GB) copied, 41.2095 s, 104 MB/s
88589934592 bytes (8.6 GB) copied, 81.5821 s, 105 MB/s

 

@hellion ref what you said, do you have an easy command I can "Copy&Paste" to avoid that "problem"?

Edited by Polanskiman
Added code tag.
Link to post
Share on other sites
  • 3 weeks later...

I did a "rebuild" of the above mentioned Asrock Q1900-ITX

DS918+ Jun's Mod V1.04b, DSM 6.2.1-23824 Update 1

2*1TB, raid0, ext4.

dd if=/dev/zero bs=1M count=1024 | md5sum
073741824 bytes (1.1 GB) copied, 3.27853 s, 328 MB/s
admin@ITX:/$ sudo dd bs=1M count=256 if=/dev/zero of=/volume1/test conv=fdatasyn
268435456 bytes (268 MB) copied, 2.01971 s, 133 MB/s
4395630592 bytes (4.4 GB) copied, 25.7595 s, 171 MB/s

I assume the improved disk speed is related to the raid0, but there is a minor decrease in CPU speed that I assume is related to the switch to DS918+.

Edited by Polanskiman
Added code tag.
Link to post
Share on other sites

esxi vm with rdm

 

vol1 - btrfs raid1

sudo dd bs=1M count=256 if=/dev/zero of=/volume1/testx conv=fdatasync
Password:
256+0 records in
256+0 records out
268435456 bytes (268 MB) copied, 2.71714 s, 98.8 MB/s

vol2 - btrfs raid0

sudo dd bs=1M count=256 if=/dev/zero of=/volume2/testx conv=fdatasync
256+0 records in
256+0 records out
268435456 bytes (268 MB) copied, 2.15798 s, 124 MB/s

and cpu XEON E3-1260l

dd if=/dev/zero bs=1M count=1024 | md5sum
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB) copied, 2.56942 s, 418 MB/s
Edited by luchuma
Added code tag.
Link to post
Share on other sites
  • 3 weeks later...

Intel(R) Pentium(R) CPU G4560 @ 3.50GHz / 8GB

DSM 6.1.7-15284

DS3617xs:~# grep MHz /proc/cpuinfo
cpu MHz         : 3500.000
cpu MHz         : 3500.000
cpu MHz         : 3500.000
cpu MHz         : 3500.000
DS3617xs:~# dd if=/dev/zero bs=1M count=1024 | md5sum
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB) copied, 1.69853 s, 632 MB/s
cd573cfaace07e7949bc0c46028904ff  -

4x 4TB SHR (LSI SAS 9211-8i )

DS3617xs:~# sudo dd bs=1M count=10240 if=/dev/zero of=/volume3/HDD_Vol3/testx conv=fdatasync
10240+0 records in
10240+0 records out
10737418240 bytes (11 GB) copied, 36.9459 s, 291 MB/s

5x5TB raid 5 (onboard)

DS3617xs:~# sudo dd bs=1M count=10240 if=/dev/zero of=/volume4/HDD_Vol4/testx conv=fdatasync
10240+0 records in
10240+0 records out
10737418240 bytes (11 GB) copied, 52.9168 s, 203 MB/s

System was scrubbing btw

Edited by Polanskiman
Added code tag.
Link to post
Share on other sites
  • 2 months later...

hello,

 

thank you very much and I think that this benchmark tool is very useful.

 

I have only problem with CPU test. In my case this  if=/dev/zero is on HDD. Speeds are same like with HDD tests. 

CPU test:

piotr@BM-918:/$ dd if=/dev/zero bs=2M count=1024 | md5sum
1024+0 records in
1024+0 records out
2147483648 bytes (2.1 GB) copied, 24.5877 s, 87.3 MB/ 

HDD test:

2684354560 bytes (2.7 GB) copied, 30.1107 s, 89.1 MB/s

 

Maybe it would be possible to run Geekbench4 directly on Xpenology?

 

Asrock J3455M

 

 

Edited by Polanskiman
Added code tag.
Link to post
Share on other sites
  • 3 weeks later...

Hp Gen 8 4x6tb raid 0(and yes i have backup server) Loader 1.02b

 

admin@HPDSMXEON:/$ sudo dd bs=1M count=32000 if=/dev/zero of=/volume1/Public/testx conv=fdatasync
32000+0 records in
32000+0 records out
33554432000 bytes (34 GB) copied, 51.8704 s, 647 MB/s

:lol:

10 gbit card is comming


XPENOLOGY makes server monster 

thank you at all for the best server os ever:17_heart_eyes:

Edited by Polanskiman
Added code tag.
Link to post
Share on other sites
  • 2 months later...

Hi Fellows,

I am new here however I have installed 2 systems one based on 1.02b mod loader using 3615XS,V6.17and another 1.04b loader using DS918+ V6.2.2 using i3 3470 and AMD A12-9800 respectively.  Both systems work fine. I am not sure what software you guys use to test the system's performance, please direct me and/or name the Syn. pkg used or using an external s/w pkgs used.  A simple direction on how to use it would help, Thanks.

Link to post
Share on other sites

My contributions:

System 1

i5-3470 using 1.02b mod loader emulating DS3615XS using DSM 6.1.7

CPU:582 Mb/s

System 2

AMD A12-9800 using 1.04b loader emulating DS918+ using DSM 6.2.2

CPU:633 Mb/s

The performance of both CPUs are comparable.

I didn't test the HDD cos it depends on the type and made/brand of the hard drives. From assessing the results above,  the CPU performance does not really improve that much if you use  i3 to i5. Celeron J3XXX or j1XXX is clearly inferior and the performance is about 1/2 of  i3 or i5.

Link to post
Share on other sites
On 3/25/2019 at 8:30 PM, Pejcik said:

hello,

 

thank you very much and I think that this benchmark tool is very useful.

 

I have only problem with CPU test. In my case this  if=/dev/zero is on HDD. Speeds are same like with HDD tests. 

CPU test:


piotr@BM-918:/$ dd if=/dev/zero bs=2M count=1024 | md5sum
1024+0 records in
1024+0 records out
2147483648 bytes (2.1 GB) copied, 24.5877 s, 87.3 MB/ 

HDD test:

2684354560 bytes (2.7 GB) copied, 30.1107 s, 89.1 MB/s

 

Maybe it would be possible to run Geekbench4 directly on Xpenology?

 

Asrock J3455M

 

 

If you are using the J3455 on Asrock the CPU performance I find is little low, I think you should be getting about CPU=255Mb/s.

Edited by Polanskiman
Added code tag.
Link to post
Share on other sites

Now I made new installation after few months but with 1 SSD. 918+ 6.2.2

Results are similar: 

/$ dd if=/dev/zero bs=2M count=1024 | md5sum
1024+0 records in
1024+0 records out
2147483648 bytes (2.1 GB) copied, 20.6168 s, 104 MB/s
/$ dd if=/dev/zero bs=10M count=1024 | md5sum  
1024+0 records in
1024+0 records out
10737418240 bytes (11 GB) copied, 80.3615 s, 134 MB/s
/$ sudo dd bs=5M count=256 if=/dev/zero of=/volume1/Share918/testx conv=fdatasync
256+0 records in
256+0 records out
1342177280 bytes (1.3 GB) copied, 4.18094 s, 321 MB/s

And CPU load is very low during that tests - maybe 17%

 

syn.png

Edited by Polanskiman
Added code tag.
Link to post
Share on other sites

J4205-ITX DS918+ Jun's Mod V1.04b

 

dd bs=1M count=1k if=/dev/zero of=/dev/null | md5sum
d41d8cd98f00b204e9800998ecf8427e  -
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB) copied, 0.164326 s, 6.5 GB/s
Edited by Polanskiman
Added code tag.
Link to post
Share on other sites
  • 1 month later...

E3 1220V2  DS3617 Jun's Mod V1.03b  DSM 6.2

CPU

# dd if=/dev/zero bs=1M count=1024 | md5sum
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB) copied, 1.80579 s, 595 MB/s
cd573cfaace07e7949bc0c46028904ff 

~# time $(i=0; while (( i < 9999999 )); do (( i ++ )); done)
real    0m39.807s
user    0m39.681s
sys     0m0.023s

 

HDD 6TBX5 Raid5

# dd bs=1M count=256 if=/dev/zero of=/volume1/share/testx conv=fdatasync
256+0 records in
256+0 records out
268435456 bytes (268 MB) copied, 0.725 s, 370 MB/s

# dd bs=1M count=1024 if=/dev/zero of=/volume1/share/testx conv=fdatasync
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB) copied, 2.27316 s, 472 MB/s

# dd bs=1M count=2048 if=/dev/zero of=/volume1/share/testx conv=fdatasync
2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB) copied, 4.19457 s, 512 MB/s
Link to post
Share on other sites
  • 2 months later...

CPU XEON E3-1270v2

 

dd if=/dev/zero bs=1M count=1024 | md5sum
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB) copied, 1.68007 s, 639 MB/s

3x WD Blue 7200 1TB Raid5

 

sudo dd bs=1M count=256 if=/dev/zero of=/volume1/testx conv=fdatasync
Password:
256+0 records in
256+0 records out
268435456 bytes (268 MB) copied, 1.52895 s, 176 MB/s

 

Link to post
Share on other sites
  • 1 month later...

Loader: jun 1.04b (DS918+)

DSM: 6.2.2-24922 Update 5

Hardware: Baremetal, Chieftec ix-03b-op, Asrock J5005 ITX, 1x8GB RAM (Single Channel), 2x5TB Seagate ST5000LM000 SHR-1

 

/$ dd if=/dev/zero bs=1M count=32768 | md5sum
32768+0 records in
32768+0 records out
34359738368 bytes (34 GB) copied, 74.3319 s, 462 MB/s
69a85eaa6fd28784c634a6bcb1d5984c  -

CPU Load durig test ~28%

/$ time $(i=0; while (( i < 9999999 )); do (( i ++ )); done)

real    0m49.016s
user    0m49.006s
sys     0m0.000s

CPU Load during test ~25%

Link to post
Share on other sites

ASROCK Q1900M + 8GB Ram + 2 HDD

DSM 6.2.2-24922 Update 4 - Loader : JUN'S LOADER v1.03b - DS3617xs + extra.lzma @IG-88 custom test 0.5 for DS3617xs

root@MiniATX:~# grep MHz /proc/cpuinfo
cpu MHz         : 1993.000
cpu MHz         : 1993.000
cpu MHz         : 1993.000
cpu MHz         : 1993.000
root@MiniATX:~# time $(i=0; while (( i < 9999999 )); do (( i ++ )); done)

real    2m16.194s
user    2m16.185s
sys     0m0.001s
root@MiniATX:~# dd if=/dev/zero bs=1M count=1024 | md5sum
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB) copied, 3.17015 s, 339 MB/s
Edited by neo82ita
Link to post
Share on other sites
  • 4 weeks later...

AsRock Z390M-ITX motherboard (dual intel nics onboard, 6x sata connectors, M.2 slot)

Intel i3 9100

8Gb RAM

256Gb Adata 256Gb SSD as cache

4 x Seagate Ironwolf 8tb drives, SHR

 

grep MHz /proc/cpuinfo

  • cpu MHz         : 3601.000
  • cpu MHz         : 3601.000
  • cpu MHz         : 3601.000
  • cpu MHz         : 3601.000

dd if=/dev/zero bs=1M count=1024 | md5sum

  • 1024+0 records in
  • 1024+0 records out
  • 1073741824 bytes (1.1 GB) copied, 1.39658 s, 769 MB/s

sudo dd bs=1M count=256 if=/dev/zero of=/volume1/Downlds/testx conv=fdatasync

  • 256+0 records in
  • 256+0 records out
  • 268435456 bytes (268 MB) copied, 0.645875 s, 416 MB/s

Tell you what - these latest core i3 CPU's rock!! I looked at building an older server, but the price of building a brand new 'low end' one outperformed most that I could see for half the price of a 1019+

Link to post
Share on other sites
  • 2 weeks later...

DSM: 6.2.2-24922 Update 5

Loader version and model: JUN'S LOADER v1.04b - DS918+

Using custom extra.lzma: 0.8_syno ds918+ by IG-88

Hardware: Baremetal,  ASRock J4105 ITX, 1x8GB RAM (Single Channel), 2x2TB Seagate IronWolf ST2000VN004 RAID 1 Btrfs

 

dd if=/dev/zero bs=1M count=1024 | md5sum
1024+0 records in
1024+0 records out
1073741824 bytes (1.1 GB) copied, 2.55064 s, 422 MB/s

 

Median of 5 tests:

sudo dd bs=1M count=256 if=/dev/zero of=/volume1/share/testx conv=fdatasync
256+0 records in
256+0 records out
268435456 bytes (268 MB) copied, 1.99444 s, 135 MB/s
Edited by LimitedModz
Link to post
Share on other sites
On 11/2/2018 at 1:51 PM, flyride said:

Hint: Repeat the test a number of times and report the median value.

 

Loader: jun 1.04b (DS918)

DSM: 6.2.1-23824U1

Hardware/CPU: J4105-ITX

HDD: WD Red 8TB RAID 5 (4 drives)

Results:


dd if=/dev/zero bs=1M count=1024 | md5sum

CPU: 422 MBps


dd bs=1M count=256 if=/dev/zero of=testx conv=fdatasync

WD Red RAID 5-4: 157 MBps

 

______________________________________________________________________________

 

My main rig is not 6.2.1, but I thought to record the results on the NVMe platform.  I'll repeat at such a date where it is converted to 6.2.1.

 

Loader: jun 1.02b (DS3615xs)

DSM: 6.1.7-15284U2

Hardware: ESXi 6.5

CPU: E3-1230v6

HDD: Intel P3500 2GB NVMe RAID 1, WD Red 4TB RAID 10 (8 drives)

Results:


dd if=/dev/zero bs=1M count=1024 | md5sum

CPU: 629MBps (I do have one other active VM but it's pretty idle)


dd bs=1M count=256 if=/dev/zero of=testx conv=fdatasync

NVMe RAID 1: 1.1 GBps

WD Red RAID 10-8: 371 MBps

 

The only thing above that can be directly compared is haldi's RAID5 @ 171MBps vs. mine at 157MBps, although the drives are quite different designs.

 

Replaced the rust-based RAID10 to a 5-way RAIDF1 using Samsung PM963 SATA SSD's.  It tests out right at 1.0 GBps.

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.