time-nuts@lists.febo.com

Discussion of precise time and frequency measurement

View all threads

Using NTP Server Time To Measure WAN Performance

FR
Fleeson, Richard
Wed, Mar 22, 2023 9:18 PM

Time Nuts

I have a question about using NTP servers to measure network performance.  I have a MPLS Wide Area Network with GPS locked NTP servers at each location.  I need to measure the network packet delay as well as packet jitter.  Looking on the NIST website I found this document via a trail of external web links:

40th Annual Precise and Time Interval (PTTI) Meeting
Studying Network Timing With Precision Packet Delay Measurements
https://apps.dtic.mil/sti/pdfs/ADA503429.pdf

Can anyone recommend a low cost or freeware way to measure packet delay and packet jitter using NTP associations or IEEE-1588?  Don't care if it is Linux or Windows OS.

Thanks,
Richard
CONFIDENTIALITY NOTICE: This e-mail may contain information that is privileged, confidential or otherwise protected from disclosure. If you are not the intended recipient of this e-mail, please notify the sender immediately by return e-mail, purge it and do not disseminate or copy it.

Time Nuts I have a question about using NTP servers to measure network performance. I have a MPLS Wide Area Network with GPS locked NTP servers at each location. I need to measure the network packet delay as well as packet jitter. Looking on the NIST website I found this document via a trail of external web links: 40th Annual Precise and Time Interval (PTTI) Meeting Studying Network Timing With Precision Packet Delay Measurements https://apps.dtic.mil/sti/pdfs/ADA503429.pdf Can anyone recommend a low cost or freeware way to measure packet delay and packet jitter using NTP associations or IEEE-1588? Don't care if it is Linux or Windows OS. Thanks, Richard CONFIDENTIALITY NOTICE: This e-mail may contain information that is privileged, confidential or otherwise protected from disclosure. If you are not the intended recipient of this e-mail, please notify the sender immediately by return e-mail, purge it and do not disseminate or copy it.
MD
Magnus Danielson
Wed, Mar 22, 2023 11:26 PM

Hi,

I know that GMV have done tests with RPi running chrony to do such tests
and reported in various forums quite recently.

Not perticular advanced tests, but that was revealing enough. They did
care to check and calibrate things thought.

They took some care to record data and present it, consider using the
influxdb and grafana combo. I typically use a python script to parse
whatever and post into influxdb.

Cheers,
Magnus

On 2023-03-22 22:18, Fleeson, Richard via time-nuts wrote:

Time Nuts

I have a question about using NTP servers to measure network performance.  I have a MPLS Wide Area Network with GPS locked NTP servers at each location.  I need to measure the network packet delay as well as packet jitter.  Looking on the NIST website I found this document via a trail of external web links:

40th Annual Precise and Time Interval (PTTI) Meeting
Studying Network Timing With Precision Packet Delay Measurements
https://apps.dtic.mil/sti/pdfs/ADA503429.pdf

Can anyone recommend a low cost or freeware way to measure packet delay and packet jitter using NTP associations or IEEE-1588?  Don't care if it is Linux or Windows OS.

Thanks,
Richard
CONFIDENTIALITY NOTICE: This e-mail may contain information that is privileged, confidential or otherwise protected from disclosure. If you are not the intended recipient of this e-mail, please notify the sender immediately by return e-mail, purge it and do not disseminate or copy it.


time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe send an email to time-nuts-leave@lists.febo.com

Hi, I know that GMV have done tests with RPi running chrony to do such tests and reported in various forums quite recently. Not perticular advanced tests, but that was revealing enough. They did care to check and calibrate things thought. They took some care to record data and present it, consider using the influxdb and grafana combo. I typically use a python script to parse whatever and post into influxdb. Cheers, Magnus On 2023-03-22 22:18, Fleeson, Richard via time-nuts wrote: > Time Nuts > > I have a question about using NTP servers to measure network performance. I have a MPLS Wide Area Network with GPS locked NTP servers at each location. I need to measure the network packet delay as well as packet jitter. Looking on the NIST website I found this document via a trail of external web links: > > 40th Annual Precise and Time Interval (PTTI) Meeting > Studying Network Timing With Precision Packet Delay Measurements > https://apps.dtic.mil/sti/pdfs/ADA503429.pdf > > Can anyone recommend a low cost or freeware way to measure packet delay and packet jitter using NTP associations or IEEE-1588? Don't care if it is Linux or Windows OS. > > Thanks, > Richard > CONFIDENTIALITY NOTICE: This e-mail may contain information that is privileged, confidential or otherwise protected from disclosure. If you are not the intended recipient of this e-mail, please notify the sender immediately by return e-mail, purge it and do not disseminate or copy it. > _______________________________________________ > time-nuts mailing list -- time-nuts@lists.febo.com > To unsubscribe send an email to time-nuts-leave@lists.febo.com
RB
Rick Bollar
Thu, Mar 23, 2023 1:40 AM

Hi Richard,

NTPViz can generate those reports and more. It works with NTPSec and
NTP and (Chrony with a

conversion script).

Sample: [1]https://time.bollar.org/plots/3hour2/#server_jitter_10.0.0.2
41

Rick

On Wed, Mar 22, 2023 at 4:18 PM, Fleeson, Richard via time-nuts
<[2]time-nuts@lists.febo.com> wrote:

 Time Nuts
 I have a question about using NTP servers to measure network
 performance. I have a MPLS Wide Area Network with GPS locked NTP
 servers at each location. I need to measure the network packet delay
 as well as packet jitter. Looking on the NIST website I found this
 document via a trail of external web links:
 40th Annual Precise and Time Interval (PTTI) Meeting
 Studying Network Timing With Precision Packet Delay Measurements
 https://apps.dtic.mil/sti/pdfs/ADA503429.pdf
 Can anyone recommend a low cost or freeware way to measure packet
 delay and packet jitter using NTP associations or IEEE-1588? Don't
 care if it is Linux or Windows OS.
 Thanks,
 Richard
 CONFIDENTIALITY NOTICE: This e-mail may contain information that is
 privileged, confidential or otherwise protected from disclosure. If
 you are not the intended recipient of this e-mail, please notify the
 sender immediately by return e-mail, purge it and do not disseminate
 or copy it.
 _______________________________________________
 time-nuts mailing list -- time-nuts@lists.febo.com
 To unsubscribe send an email to time-nuts-leave@lists.febo.com

References

  1. https://time.bollar.org/plots/3hour2/#server_jitter_10.0.0.241
  2. mailto:time-nuts@lists.febo.com
Hi Richard, NTPViz can generate those reports and more. It works with NTPSec and NTP and (Chrony with a conversion script). Sample: [1]https://time.bollar.org/plots/3hour2/#server_jitter_10.0.0.2 41 Rick On Wed, Mar 22, 2023 at 4:18 PM, Fleeson, Richard via time-nuts <[2]time-nuts@lists.febo.com> wrote: Time Nuts I have a question about using NTP servers to measure network performance. I have a MPLS Wide Area Network with GPS locked NTP servers at each location. I need to measure the network packet delay as well as packet jitter. Looking on the NIST website I found this document via a trail of external web links: 40th Annual Precise and Time Interval (PTTI) Meeting Studying Network Timing With Precision Packet Delay Measurements https://apps.dtic.mil/sti/pdfs/ADA503429.pdf Can anyone recommend a low cost or freeware way to measure packet delay and packet jitter using NTP associations or IEEE-1588? Don't care if it is Linux or Windows OS. Thanks, Richard CONFIDENTIALITY NOTICE: This e-mail may contain information that is privileged, confidential or otherwise protected from disclosure. If you are not the intended recipient of this e-mail, please notify the sender immediately by return e-mail, purge it and do not disseminate or copy it. _______________________________________________ time-nuts mailing list -- time-nuts@lists.febo.com To unsubscribe send an email to time-nuts-leave@lists.febo.com References 1. https://time.bollar.org/plots/3hour2/#server_jitter_10.0.0.241 2. mailto:time-nuts@lists.febo.com
KT
Keenan Tims
Thu, Mar 23, 2023 5:50 PM

I have used irtt[1] to make similar measurements. Even with
'semi-precision' (NTP) time sync, the one-way delay estimates and one-way
loss measurements are quite useful to pin down issues. However it has been
designed with precision timing in mind, so with PTP or stratum 1 time sync,
it will enable accurate one-way delay and jitter measurement as well.

There is no API, but the command-line client can output JSON which makes it
pretty easy to integrate into other tooling.

[1] https://github.com/heistp/irtt

On Wed, 22 Mar 2023 at 15:05, Fleeson, Richard via time-nuts <
time-nuts@lists.febo.com> wrote:

Time Nuts

I have a question about using NTP servers to measure network performance.
I have a MPLS Wide Area Network with GPS locked NTP servers at each
location.  I need to measure the network packet delay as well as packet
jitter.  Looking on the NIST website I found this document via a trail of
external web links:

40th Annual Precise and Time Interval (PTTI) Meeting
Studying Network Timing With Precision Packet Delay Measurements
https://apps.dtic.mil/sti/pdfs/ADA503429.pdf

Can anyone recommend a low cost or freeware way to measure packet delay
and packet jitter using NTP associations or IEEE-1588?  Don't care if it is
Linux or Windows OS.

Thanks,
Richard
CONFIDENTIALITY NOTICE: This e-mail may contain information that is
privileged, confidential or otherwise protected from disclosure. If you are
not the intended recipient of this e-mail, please notify the sender
immediately by return e-mail, purge it and do not disseminate or copy it.


time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe send an email to time-nuts-leave@lists.febo.com

I have used irtt[1] to make similar measurements. Even with 'semi-precision' (NTP) time sync, the one-way delay estimates and one-way loss measurements are quite useful to pin down issues. However it has been designed with precision timing in mind, so with PTP or stratum 1 time sync, it will enable accurate one-way delay and jitter measurement as well. There is no API, but the command-line client can output JSON which makes it pretty easy to integrate into other tooling. [1] https://github.com/heistp/irtt On Wed, 22 Mar 2023 at 15:05, Fleeson, Richard via time-nuts < time-nuts@lists.febo.com> wrote: > Time Nuts > > I have a question about using NTP servers to measure network performance. > I have a MPLS Wide Area Network with GPS locked NTP servers at each > location. I need to measure the network packet delay as well as packet > jitter. Looking on the NIST website I found this document via a trail of > external web links: > > 40th Annual Precise and Time Interval (PTTI) Meeting > Studying Network Timing With Precision Packet Delay Measurements > https://apps.dtic.mil/sti/pdfs/ADA503429.pdf > > Can anyone recommend a low cost or freeware way to measure packet delay > and packet jitter using NTP associations or IEEE-1588? Don't care if it is > Linux or Windows OS. > > Thanks, > Richard > CONFIDENTIALITY NOTICE: This e-mail may contain information that is > privileged, confidential or otherwise protected from disclosure. If you are > not the intended recipient of this e-mail, please notify the sender > immediately by return e-mail, purge it and do not disseminate or copy it. > _______________________________________________ > time-nuts mailing list -- time-nuts@lists.febo.com > To unsubscribe send an email to time-nuts-leave@lists.febo.com >
T
Time-Nuts
Fri, Mar 24, 2023 10:40 PM

Do your devices support RFC 2544 or Y.1564? Or does that not get you what you need?

Paul

March 22, 2023 5:18 PM, "Fleeson, Richard via time-nuts" time-nuts@lists.febo.com wrote:

Time Nuts

I have a question about using NTP servers to measure network performance. I have a MPLS Wide Area
Network with GPS locked NTP servers at each location. I need to measure the network packet delay as
well as packet jitter. Looking on the NIST website I found this document via a trail of external
web links:

40th Annual Precise and Time Interval (PTTI) Meeting
Studying Network Timing With Precision Packet Delay Measurements
https://apps.dtic.mil/sti/pdfs/ADA503429.pdf

Can anyone recommend a low cost or freeware way to measure packet delay and packet jitter using NTP
associations or IEEE-1588? Don't care if it is Linux or Windows OS.

Thanks,
Richard
CONFIDENTIALITY NOTICE: This e-mail may contain information that is privileged, confidential or
otherwise protected from disclosure. If you are not the intended recipient of this e-mail, please
notify the sender immediately by return e-mail, purge it and do not disseminate or copy it.


time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe send an email to time-nuts-leave@lists.febo.com

Do your devices support RFC 2544 or Y.1564? Or does that not get you what you need? Paul March 22, 2023 5:18 PM, "Fleeson, Richard via time-nuts" <time-nuts@lists.febo.com> wrote: > Time Nuts > > I have a question about using NTP servers to measure network performance. I have a MPLS Wide Area > Network with GPS locked NTP servers at each location. I need to measure the network packet delay as > well as packet jitter. Looking on the NIST website I found this document via a trail of external > web links: > > 40th Annual Precise and Time Interval (PTTI) Meeting > Studying Network Timing With Precision Packet Delay Measurements > https://apps.dtic.mil/sti/pdfs/ADA503429.pdf > > Can anyone recommend a low cost or freeware way to measure packet delay and packet jitter using NTP > associations or IEEE-1588? Don't care if it is Linux or Windows OS. > > Thanks, > Richard > CONFIDENTIALITY NOTICE: This e-mail may contain information that is privileged, confidential or > otherwise protected from disclosure. If you are not the intended recipient of this e-mail, please > notify the sender immediately by return e-mail, purge it and do not disseminate or copy it. > _______________________________________________ > time-nuts mailing list -- time-nuts@lists.febo.com > To unsubscribe send an email to time-nuts-leave@lists.febo.com