General list for discussing Bufferbloat
 help / color / mirror / Atom feed
From: Hal Murray <halmurray+bufferbloat@sonic.net>
To: bloat@lists.bufferbloat.net
Cc: Hal Murray <halmurray+bufferbloat@sonic.net>
Subject: [Bloat] Servers, time
Date: Mon, 12 Dec 2022 20:32:15 -0800	[thread overview]
Message-ID: <20221213043215.7EBB728C206@107-137-68-211.lightspeed.sntcca.sbcglobal.net> (raw)


[I work on NTP software so I'm interested in network delays.]

> I've been holding off on iperf 2 public servers until I found an additional
> value add and a way to pay for them.

AT&T ran fiber down my street.  I can get a gigabit link, symmetric, no data 
caps, for $80/month.  I'll be happy to setup a server if that will help.  I'm 
in Silicon Valley.

>                                                                     Much of the iperf 2 work has been
> around one way delay (OWD) or latency. Doing this well requires GPS clock
> sync on both the data center servers and the end host devices

How accurate do you need the time?

> I checked into this a few years ago and found that this level of clock sync
> wasn't available via rented servers (e.g. linode or Hurricane Electric) so I
> put on hold any further investigation of public servers for iperf 2 as being
> redundant with iperf 3. Those that need true e2e latency (vs RTTs) have to
> build their own so-to-speak. 

I can see 2 approaches.

One is to find good local NTP servers.

HE has NTP servers.
  https://www.he.net/adm/ntp.html

NIST has servers in Gaithersburg MD and Boulder and Fort Colins in Colorado.
  https://tf.nist.gov/tf-cgi/servers.cgi

Google and Cloudflare have good servers and they probably have good 
connectivity.
  https://developers.google.com/time
  https://www.cloudflare.com/time/

In Germany, PTB has good servers:
  https://www.ptb.de/cms/en/ptb/fachabteilungen/abtq/gruppe-q4/ref-q42/time-syn
chronization-of-computers-using-the-network-time-protocol-ntp.html

.....

The other approach would be to do your own NTP like time sync with the server 
before and after a test run.  This assumes you can find a quiet time to get a 
good measurement.  NTP assumes that the routing delays are symmetric.  Unless 
something wild is going on, that's a reasonable assumption.  Less so on longer 
paths.

I'll be glad to help set something up and/or run some tests.


-- 
These are my opinions.  I hate spam.




                 reply	other threads:[~2022-12-13  4:32 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

  List information: https://lists.bufferbloat.net/postorius/lists/bloat.lists.bufferbloat.net/

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20221213043215.7EBB728C206@107-137-68-211.lightspeed.sntcca.sbcglobal.net \
    --to=halmurray+bufferbloat@sonic.net \
    --cc=bloat@lists.bufferbloat.net \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox