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

Re: your mail



Hi Chris, 

I am taking this info from the LDM 5 Workshop notebook..

Feed            Volume
----            ------

DIFAX           19Mb/hour
FSL2            2Mb/hour
HDS             180Mb/hour
IDS|DDPLUS      6Mb/hour
MCIDAS          7Mb/hour
NEXRAD          73MB/hour
NLDN            291K/hour
NMC3(Floater)   3.5Mb/hour
NOGAPS          195Mb/hour


Your mileage may vary, but this should ballpark you..These are "max"
values per hour..

-Jeff
____________________________                  _____________________
Jeff Weber                                    address@hidden
Unidata Support                               PH:303-497-8676 
COMET Case Study Library                      FX:303-497-8690
University Corp for Atmospheric Research      3300 Mitchell Ln
http://www.unidata.ucar.edu/staff/jweber      Boulder,Co 80307-3000
________________________________________      ______________________

On Tue, 15 Oct 2002, Chris Herbster wrote:

> 
> 
> Jeff Weber wrote:
> 
> > Hi Chris,
> >
> > There is no need to contact the site managers at this time (diagnostics)
> > obviously if we choose to request data we will need to talk with them :)
> >
> 
> I just had a good chat with Curtis.  He's first going to upgrade to the 
> latest releases of everything and then make some timing tests with more local 
> machines.  We did one test on the phone that showed a slower communication to 
> us than to at least one other site....
> 
> 
> >
> > Sitelist:
> > http://www.unidata.ucar.edu/projects/idd/sitelist.html
> >
> > I am not quite sure what you mean by "minimum performance stats" but you
> > can modify the packet size in the ping utility to more closely imitate LDM
> > packets and their size...
> >
> 
> I know that the inverse of one of the stats (ldmping) represents the most 
> transactions that one could make (per second) with that machine.  Obviously 
> your need for transactions is a function of the volume of info you request, 
> so I was just wondering if there is a minimum level of connectivity that is 
> considered a threshold for LDM to be reliable?  (This minimum is tied to 
> volume, so you might not be able to answer without us first considering the 
> total feed request.)
> 
> I've tried to interpret the feed volumes on the graphs, but most seem to be 
> off the scale.  Can you give some estimates for the daily volumes for the 
> different feed types?
> 
> 
> >
> > Let me know how I can help if needed..
> >
> > Thanks,
> >
> > -Jeff
> > ____________________________                  _____________________
> 
> Thanks!!
> 
> CH
> 
> 
>