This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.
Hi Doug, re: > Thanks for the info! No worries. > Does the attached sketch accurately depict what you've described? Yes, but I reiterate that you might be forced into running more collector machines as there are more and more modeling centers sending ensemble data to the archive centers. One of the tuning objectives we learned in our pre-cluster investigations that it is important to try and _minimize_ the number of requests being made on a machine since this minimizes write lock contention in the LDM queue. At the same time, of course, one needs to increase the number of requests in order to minimize the latency seen on received products. In a steady state system (a system where the number of requests for data and requests to send data does not change much), one can adjust the number of data requests for maximum system performance fairly quickly. Cheers, Tom **************************************************************************** Unidata User Support UCAR Unidata Program (303) 497-8642 P.O. Box 3000 address@hidden Boulder, CO 80307 ---------------------------------------------------------------------------- Unidata HomePage http://www.unidata.ucar.edu **************************************************************************** Ticket Details =================== Ticket ID: XLU-757491 Department: Support IDD TIGGE Priority: Normal Status: Closed