My bottleneck is in the time to fetch the page.  I sync about 30-40 pages
for my company's conduit which takes about 60 - 90 seconds.  I was thinking
that fetching 2 - 3 pages at a time would cut the time in half, but this
was more about running the spider.process 2 or 3 at a time.

Bill

Your mouse has moved. Windows has to reboot for changes to take effect.


                                                                                       
                      
                                                                                       
                      
                                           To:     "Plucker Development List"          
                      
                    "Robert                 <[EMAIL PROTECTED]>            
                      
                    O'Connor"              cc:     (bcc: Bill Nalen/Towers Perrin)     
                      
                    <rob@medicalmnem       Subject:     RE: Plucker conduit -- Syncing 
multiple 'units'      
                    onics.com>              simultaneously                             
                      
                                                                                       
                      
                    10/24/2001 03:39                                                   
                      
                    PM                                                                 
                      
                    Please respond                                                     
                      
                    to Plucker                                                         
                      
                    Development List                                                   
                      
                                                                                       
                      
                                                                                       
                      




> Adding multi-process fetching of Web pages would indeed speed things
> up, but it would also increase the likelihood of bugs significantly.

I was wondering, where exactly is the usual bottleneck in the Parser in
terms of parsing speed?

I was interested in this, since was wondering if there would be a net
increase in speed if a manager was allowed to sync, say 2 or 3
(channels|databases|streams) simultaneously (ie run three instances of the
parser outputting 3 separate.pdb files).

Best wishes,
Robert





Reply via email to