[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: 20001002: pqsurf in ldm5.1.2

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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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.

  • Subject: Re: 20001002: pqsurf in ldm5.1.2
  • Date: Mon, 02 Oct 2000 14:16:00 -0600
David,

> btw Russ,
>     So far it seems that 5.1.2 is *much* more efficient
> with disk io. iowait and diskwaits are much shorter now
> than they have been in the past on our ldm machine.
> Thanks

Glad to hear it, thanks for the feedback.  So far only about 36% of
our LDM sites have converted to 5.1.2, and very few have provided any
comments on any differences they have noticed.  I'm trying to put
together a couple of graphs for the PolComm meeting on the data rate in
products per second with 5.0 versus 5.1, depending on the number of
products in the queue.  This one summarizes things:

      Table 1. Measured rates of typical
   product insertion in a full product queue

   Products  Products/sec,  Products/sec,
   in Queue     LDM 5.0        LDM 5.1

      20000      130            300

      40000       66            243

      80000       40            230

     160000       20            210

--Russ