ATTENDEES:
Till, Tom, Terri, Stephanie, Debbie, Qing, Sonya, Ernest, Kukhee, Diane
NOTES:
My notes are pretty cryptic. Please feel free to email additions or add to the wiki page.
suggestion - it would be great to get the diagrams with even greater detail, including EVR interface, MPS, BSA
suggestion - we should eventually get all diagnostic devices onto this network, could be used as source for BSA
the room agreed that the function-ptr table was the best option for choosing the 'flavor' of udpComm for each bpm ioc.
Steph and Till have a couple BSA related items to discuss off-line. Changes to the EVR modifiers and possible ways to speed up BSA processing. Just a reminder here...
Till pointed out that in the current design, if a BPM does not work, then the entire packet of BPM data will not be sent. This may not be much of a problem for feedbacks but on further thought it would not be very good for most other use cases - BSA for example. It is worth the time to try to devise a not-to-difficult scheme that will allow the data from the working BPMs to be delivered.
Till or Sonya needs to set the timestamp TSE = -2 or something like that...
It was agreed that the subnet should be called FNET (Fast Net; lclsfnet) and DNS names for FCOM should use -fnet, not -fbck, since this subnet may be used for more than just fast feedback.
ACTION ITEMS:
FCOM IDs and blobs
DNS names for 2nd NIC
Status
Mcast ID for vme-coldfire communications
udpComm