NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.
Current content can be found at unidata.ucar.edu.
To learn about what's going on, see About the Archive Site.
NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.
Current content can be found at unidata.ucar.edu.
To learn about what's going on, see About the Archive Site.
> > NCF maximum latency . . . the time the product spends on queue at the NCF > uplink . . . was 62 seconds since 2 Feb. Average has been 2 seconds on the > NCF NWSTG queue. This is quite good . . . especially considering latencies > on EMWIN is upwards of 15 minutes for "emergency" products, and hours for > imagery. Also, since the first of the year, the NWSTG traffic has exploded > from a little over 4GB per day to 5.9GB per day (O.K. - let's just call it > 6GB per day) - 50% increase. > > Amount of time a product spends on a "queue" after downlink is vendor and > location specific. > ---------- > > IDD latency will undoubtedly be landline (internet) or IDD receive system > related. Emwin uses a priority queueing system. Severe and Tornado warnings get queued in front of everything. Imagery and graphics get dead last and only get sent after everything else, IF there is time. The traffic rate is highly variable. Graphics often get dumped since they are so much larger than text products. 15 mins would effectively render emwin down. TV, radio, Police, and numerous emergency management agencies utilize emwin to receive real time warnings. Ive never seen a 15 min queue on warnings. The whole purpose of emwin is to enable fast low cost disemmination of warnings to emergency personell. I don't think its a good example to compare latency against due to the queuing. Ray Weber MA Skywarn
ldm-users archives: