[ntp:questions] NTP tunning for OWD measurements

E-Mail Sent to this address will be added to the BlackLists Null at BlackList.Anitech-Systems.invalid
Sat Oct 27 02:04:27 UTC 2012

John Hasler wrote:
> Read this: <http://www.internet2.edu/performance/owamp/>.
>  I think he wants help configuring the four clocks
>  mentioned in the "requirements" optimally for OWAMP.
>  I don't see why OWAMP is special, though.

OWAMP doesn't appear to mean without a e.g. PPS source,
  or perhaps local LAN to a Stratum 1 with a PPS source,
  certainly not ntpd synced only via routed connections,
  if they want high accuracy and precision,
  as micro seconds appear to be a issue?

Effect Of ATOM PPS Driver Edge Change on OWAMP Data
 The graph above shows one-way transit times as measured by owamp.
  The two hosts are separated by a single Gigabit Ethernet switch,
   and at about 11:30 AM, NTP's ATOM refclock was reconfigured
   with “fudge flag1 1”, which supposedly uses the falling edge.
  This graph suggests, however, that the change has caused the
   PPS-configured host ( to alter its time to 104 μS later,
   which suggests that the trailing/rising edge is used when
   “fudge flag2 1” is configured.
  The fact that this would align the time with the beginning
   of the Truetime CR start bit further supports that the ATOM
   PPS driver is using the trailing/rising edge of the PPS pulse.

HOTS: An OWAMP-Compliant Hardware Packet Timestamp Recorder
 "accepts an external 10-MHz signal as well as 1PPS signal as input"
 two mini-BNC jacks are used to obtain clock information
  from an external source such as a GPS or CDMA receiver.
 HOTS can accept two kinds of signal as input: a 1PPS
  signal and a 10-MHz signal.
 The precision of the generated timestamp depends on the
  accuracy of the provided signals.
 In our measurements, we used two kinds of GPS receiver
  to generate these signals: the HP 58503A and the TymServe
 Both of them can generate extremely precise clock information.
Highly precise OWD measurement is a challenge because of
  the difficulty of eliminating errors in process to obtain
  an accurate timestamp and other overheads.
 In this paper, we introduced HOTS - a hardware packet
  timestamper that we developed to measure OWD.
 HOTS can generate extremely precise clock information
  for OWAMP test packets provided an accurate time source
  such as a GPS or CDMA receiver.

RFC4656 A One-way Active Measurement Protocol (OWAMP)
 High-precision measurement of these one-way IP performance
  metrics became possible with wider availability of good
  time sources (such as GPS and CDMA).
 With the increasingly wide availability of affordable
  global positioning systems (GPS) and CDMA-based time
  sources, hosts increasingly have available to them very
  accurate time sources, either directly or through their
  proximity to Network Time Protocol (NTP) primary
  (stratum 1) time servers.
 4.1.2. OWAMP-Test Packet Format and Content
 The first bit, S, SHOULD be set if the party generating
   the timestamp has a clock that is synchronized to UTC
   using an external source (e.g., the bit should be set
   if GPS hardware is used and it indicates that it has
   acquired current position and time or if NTP is used
   and it indicates that it has synchronized to an external
   source, which includes stratum 0 source, etc.).
  If there is no notion of external synchronization for
   the time source, the bit SHOULD NOT be set.

Some other threads I read on the subject had people
 complaining about issues with too much jitter from their
 GPS affecting their attempts at OWAMP measurements in the
 10s of microseconds.

To answer the OP's original question, here is a reference:
#server prefer       # 5 is trimble TRUETIME/ 0 is ttyd0
#fudge refid CDMA time1 0.013
#fudge refid CDMA

server prefer       # 5 is trimble TRUETIME/ 0 is ttyd0
fudge refid CDMA

server minpoll 2 maxpoll 6 # 22 is PPS "ATOM" device
fudge flag2 1 flag3 1

# nms4-chin
# nms3-atla
# nms4-kscy
# tock.cerias.purdue.edu

E-Mail Sent to this address <BlackList at Anitech-Systems.com>
  will be added to the BlackLists.

More information about the questions mailing list