Hi, You using Telecom profiles?
The traditional way to do this is to use a dedicated time tester to do that. The most deployed (in my world, anyway) is the Calnex Paragon-X. This device does T-BC testing using 2 ports with master + slave emulation. The Paragon tester is then connected back-to-back to the T-BC. It has a master port which feeds PTP (+SyncE) to the T-BC and it measures the returned signal in the slave port. Then there is a post-processing analysis tool (CAT) which generated the statistics. It seems you are interested in the “Noise Generation” test. This is a standardized test for a T-BC specified in G.8273.2 (I think it’s section 7.1). [cid:image001.png@01D66B51.F37CBE50] Hope that helps, Dennis Hagarty Cisco Systems. From: Prashanth Candade <prash.cand...@gmail.com> Sent: Wednesday, 05 August 2020 12:42 To: linuxptp-users@lists.sourceforge.net Subject: [Linuxptp-users] Metrics: Constant Time Error and Dynamic Time Error Hi, We have an implementation of Boundary Clock based on ptp4l and phc2sys. This is on a CentOS box. Scenarios we support, BC has one slave and one Master BC has one slave and 3-4 Masters As part of metrics for PTP performance, we need to provide the following data -> Constant Time Error -> Dynamic Time Error Both according to G.8260 definitions. Can you please recommend how we can calculate these for our implementation of BC? Thanks, Prashanth
_______________________________________________ Linuxptp-users mailing list Linuxptp-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/linuxptp-users
