-
Type: Task
-
Status: Done (View Workflow)
-
Priority: Normal
-
Resolution: Done
-
Component/s: None
-
Labels:None
These days, Linux includes decent PTP support, and PTP can be used in place of NTP. Do we want to? The underlying question is whether we will ever need instrument-wide timing better than a few milliseconds.
The one obvious problem which could benefit from precise relative instrument-wide timing is matching fiber motion or LED strobes to MCS rows. If there is any chance we will want to synchronize those I believe we should use PTP. If not, NTP is fine.
I don't think there is a significant difference in management effort these days. The head server can use the observatory NTP feed for its absolute time, regardless of the chosen protocol.
- blocks
-
INSTRM-291 Provide development computer(s) at Subaru
- Done