Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Bandwidth measurements are not accurate #171

Open
fanqiehc opened this issue Feb 10, 2022 · 4 comments
Open

Bandwidth measurements are not accurate #171

fanqiehc opened this issue Feb 10, 2022 · 4 comments

Comments

@fanqiehc
Copy link

The bandwidth gap between server and client is around 10% .


# ./ethr -s -4

Ethr: Comprehensive Network Performance Measurement Tool (Version: v1.0.0)
Maintainer: Pankaj Garg (ipankajg @ LinkedIn | GitHub | Gmail | Twitter)

-----------------------------------------------------------
Accepting IP version: ipv4
Listening on port 8888 for TCP & UDP
-----------------------------------------------------------
[RemoteAddress]  Proto   Bits/s   Conn/s    Pkt/s   Latency
[    10.3.9.21]    TCP  265.09M        1     --        --
[    10.3.9.21]    TCP    9.60G        0     --        --
[    10.3.9.21]    TCP    9.07G        0     --        --
[    10.3.9.21]    TCP    9.09G        0     --        --
[    10.3.9.21]    TCP    9.66G        0     --        --
[    10.3.9.21]    TCP    9.48G        0     --        --
[    10.3.9.21]    TCP    9.47G        0     --        --
[    10.3.9.21]    TCP    9.61G        0     --        --
[    10.3.9.21]    TCP    9.19G        0     --        --
[    10.3.9.21]    TCP    9.16G        0     --        --
[    10.3.9.21]    TCP    9.55G        0     --        --


./ethr -c 10.3.9.14

Ethr: Comprehensive Network Performance Measurement Tool (Version: v1.0.0)
Maintainer: Pankaj Garg (ipankajg @ LinkedIn | GitHub | Gmail | Twitter)

Using destination: 10.3.9.14, ip: 10.3.9.14, port: 8888
[  5] local 10.3.9.21 port 35617 connected to 10.3.9.14 port 8888
- - - - - - - - - - - - - - - - - - - - - - -
[  ID ]   Protocol    Interval      Bits/s
[    5]     TCP      000-001 sec     8.73G
[    5]     TCP      001-002 sec     8.34G
[    5]     TCP      002-003 sec     8.33G
[    5]     TCP      003-004 sec     8.64G
[    5]     TCP      004-005 sec     8.55G
[    5]     TCP      005-006 sec     8.56G
[    5]     TCP      006-007 sec     8.61G
[    5]     TCP      007-008 sec     8.43G
[    5]     TCP      008-009 sec     8.37G
[    5]     TCP      009-010 sec     8.57G
Ethr done, duration: 10s.
Hint: Use -d parameter to change duration of the test.
@pgmsft
Copy link
Collaborator

pgmsft commented Mar 14, 2022

Could you please share latency information between server and client? I will try to reproduce this problem on my side.

@difficultwork
Copy link

difficultwork commented Jun 21, 2022 •
edited
Loading

@pgmsft
I have same problem. Should I trust server or client？

./ethr -s -port 8080

[RemoteAddress]  Proto   Bits/s   Conn/s    Pkt/s   Latency
[  192.168.2.2]    TCP  627.20M       64     --        --  
[  192.168.2.2]    TCP    6.85G        0     --        --  
[  192.168.2.2]    TCP    7.87G        0     --        --  
[  192.168.2.2]    TCP    7.66G        0     --        --  
[  192.168.2.2]    TCP    7.72G        0     --        --  
[  192.168.2.2]    TCP    7.72G        0     --        --  
[  192.168.2.2]    TCP    7.82G        0     --        --  
[  192.168.2.2]    TCP    8.21G        0     --        --  
[  192.168.2.2]    TCP    7.92G        0     --        --  
[  192.168.2.2]    TCP    7.89G        0     --        --

./ethr -c 192.168.1.3 -port 8080 -p tcp -n 64

- - - - - - - - - - - - - - - - - - - - - - -
[    5]     TCP      009-010 sec    15.49M
[    6]     TCP      009-010 sec    15.62M
[    7]     TCP      009-010 sec    14.85M
[    8]     TCP      009-010 sec    16.77M
[    9]     TCP      009-010 sec    14.08M
[   10]     TCP      009-010 sec    17.41M
[   11]     TCP      009-010 sec    14.08M
[   12]     TCP      009-010 sec    16.13M
[   13]     TCP      009-010 sec    16.77M
[   14]     TCP      009-010 sec    16.77M
[   15]     TCP      009-010 sec    15.49M
[   16]     TCP      009-010 sec    16.13M
[   17]     TCP      009-010 sec    16.13M
[   18]     TCP      009-010 sec    16.64M
[   19]     TCP      009-010 sec    15.49M
[   20]     TCP      009-010 sec    15.49M
[   21]     TCP      009-010 sec    16.26M
[   22]     TCP      009-010 sec    19.33M
[   23]     TCP      009-010 sec    15.49M
[   24]     TCP      009-010 sec    14.85M
[   25]     TCP      009-010 sec    17.41M
[   26]     TCP      009-010 sec    15.36M
[   27]     TCP      009-010 sec    16.77M
[   28]     TCP      009-010 sec    15.49M
[   29]     TCP      009-010 sec    18.18M
[   30]     TCP      009-010 sec    16.13M
[   31]     TCP      009-010 sec    17.54M
[   32]     TCP      009-010 sec    16.77M
[   33]     TCP      009-010 sec    15.49M
[   34]     TCP      009-010 sec    14.21M
[   35]     TCP      009-010 sec    17.41M
[   36]     TCP      009-010 sec    15.62M
[   37]     TCP      009-010 sec    14.85M
[   38]     TCP      009-010 sec       16M
[   39]     TCP      009-010 sec    14.21M
[   40]     TCP      009-010 sec    15.49M
[   41]     TCP      009-010 sec       16M
[   42]     TCP      009-010 sec    14.98M
[   43]     TCP      009-010 sec    16.13M
[   44]     TCP      009-010 sec    14.85M
[   45]     TCP      009-010 sec    16.77M
[   46]     TCP      009-010 sec    15.62M
[   47]     TCP      009-010 sec    15.62M
[   48]     TCP      009-010 sec       16M
[   49]     TCP      009-010 sec    15.49M
[   50]     TCP      009-010 sec    15.49M
[   51]     TCP      009-010 sec    13.44M
[   52]     TCP      009-010 sec    15.49M
[   53]     TCP      009-010 sec    15.49M
[   54]     TCP      009-010 sec    14.85M
[   55]     TCP      009-010 sec    17.41M
[   56]     TCP      009-010 sec    16.64M
[   57]     TCP      009-010 sec    16.13M
[   58]     TCP      009-010 sec    15.36M
[   59]     TCP      009-010 sec    16.77M
[   60]     TCP      009-010 sec    15.49M
[   61]     TCP      009-010 sec    15.49M
[   62]     TCP      009-010 sec       16M
[   63]     TCP      009-010 sec    15.49M
[   64]     TCP      009-010 sec    15.49M
[   65]     TCP      009-010 sec       16M
[   66]     TCP      009-010 sec    15.36M
[   67]     TCP      009-010 sec    13.57M
[   68]     TCP      009-010 sec    15.49M
[  SUM]     TCP      009-010 sec     1.01G
- - - - - - - - - - - - - - - - - - - - - - -

@AvrumFeldman
Copy link

AvrumFeldman commented Aug 10, 2022 •
edited
Loading

I have the same issue.

Client side

ethr.exe -c ... -n 0
....
Using destination: ..., ip: 192.168.200.68, port: 8888
...
- - - - - - - - - - - - - - - - - - - - - - -
[  ID ]   Protocol    Interval      Bits/s
...
[  SUM]     TCP      000-001 sec     8.45G
- - - - - - - - - - - - - - - - - - - - - - -
...
[  SUM]     TCP      001-002 sec    10.23G
- - - - - - - - - - - - - - - - - - - - - - -
...
[  SUM]     TCP      002-003 sec     9.65G
- - - - - - - - - - - - - - - - - - - - - - -
...
[  SUM]     TCP      003-004 sec     9.60G
- - - - - - - - - - - - - - - - - - - - - - -
...
[  SUM]     TCP      004-005 sec     9.47G
- - - - - - - - - - - - - - - - - - - - - - -
...
[  SUM]     TCP      005-006 sec     9.54G
- - - - - - - - - - - - - - - - - - - - - - -
...
[  SUM]     TCP      006-007 sec     9.54G
- - - - - - - - - - - - - - - - - - - - - - -
...
[  SUM]     TCP      007-008 sec     9.49G
- - - - - - - - - - - - - - - - - - - - - - -
...
[  SUM]     TCP      008-009 sec     9.50G
- - - - - - - - - - - - - - - - - - - - - - -
....
[  SUM]     TCP      009-010 sec     9.49G
- - - - - - - - - - - - - - - - - - - - - - -
Ethr done, duration: 10s.

But on server side I get quite a different number

./ethr -s

[RemoteAddress]  Proto   Bits/s   Conn/s    Pkt/s   Latency
[...168.200.72]    TCP   13.57M        4     --        --
[...168.200.72]    TCP    8.20G       45     --        --
[...168.200.72]    TCP   11.47G       15     --        --
[...168.200.72]    TCP   11.83G        0     --        --
[...168.200.72]    TCP   11.83G        0     --        --
[...168.200.72]    TCP   12.62G        0     --        --
[...168.200.72]    TCP   13.34G        0     --        --
[...168.200.72]    TCP   13.89G        0     --        --
[...168.200.72]    TCP      14G        0     --        --
[...168.200.72]    TCP   14.24G        0     --        --
[...168.200.72]    TCP   14.35G        0     --        --

Following is a latency test between the same two hosts

ethr.exe -c ... -n 0 -t l

...

Using destination: ..., ip: 192.168.200.68, port: 8888
Running latency test: 1000, 1
-----------------------------------------------------------------------------------------
      Avg       Min       50%       90%       95%       99%     99.9%    99.99%       Max
  1.194ms   0.000ns   1.420ms   1.890ms   2.002ms   2.548ms   3.013ms   3.013ms   3.045ms
432.270us   0.000ns 369.700us   1.042ms   1.334ms   1.431ms   1.465ms   1.465ms   1.492ms
441.727us   0.000ns 379.100us   1.057ms   1.362ms   1.430ms   1.457ms   1.457ms   1.466ms
526.781us   0.000ns 401.800us   1.312ms   1.361ms   1.422ms   1.468ms   1.468ms   1.471ms
514.106us   0.000ns 399.300us   1.297ms   1.383ms   1.435ms   1.454ms   1.454ms   1.477ms
471.030us   0.000ns 396.600us   1.030ms   1.343ms   1.449ms   1.488ms   1.488ms   1.489ms
421.487us   0.000ns 337.200us   1.030ms   1.327ms   1.423ms   1.465ms   1.465ms   1.474ms
516.162us   0.000ns 403.400us   1.300ms   1.345ms   1.420ms   1.460ms   1.460ms   1.516ms
497.944us   0.000ns 395.200us   1.040ms   1.340ms   1.415ms   1.434ms   1.434ms   1.462ms
499.915us   0.000ns 393.400us   1.262ms   1.331ms   1.420ms   1.447ms   1.447ms   1.458ms
Ethr done, duration: 10s.
Hint: Use -d parameter to change duration of the test.

[RemoteAddress]  Proto   Bits/s   Conn/s    Pkt/s   Latency
[...168.200.72]    TCP        0        1     --        --
[...168.200.72]    TCP        0        0     --    431.980us
[...168.200.72]    TCP        0        0     --    438.978us
[...168.200.72]    TCP        0        0     --    527.000us
[...168.200.72]    TCP        0        0     --    514.018us
[...168.200.72]    TCP        0        0     --    471.000us
[...168.200.72]    TCP        0        0     --    422.000us
[...168.200.72]    TCP        0        0     --    516.036us
[...168.200.72]    TCP        0        0     --    497.002us
[...168.200.72]    TCP        0        0     --    499.980us

In this case I believe the client speeds are more accurate as I know I have 10 Gigabit network and the clients speed is ~10 while the server is reporting ~15Gigabits/ps

@wolruf
Copy link

wolruf commented Aug 16, 2022

you may want to try this patch: eb6df4d it fixed an inaccurate bandwidth reporting in my case.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
5 participants
@wolruf @fanqiehc @difficultwork @AvrumFeldman @pgmsft