From: barmar@alum.mit.edu   
      
   In article ,   
    Glen Herrmannsfeldt  wrote:   
      
   > Barry Margolin wrote:   
   > (snip)   
   >   
   > > Why are you sending it in small chunks?  Send it in the biggest chunks   
   > > possible, to reduce the header overhead.   
   >   
   > Big and small are relative, and the OP didn't give a   
   > scale to the problem.  Later posts indicate that it   
   > is time sensitive market data.  It might be that   
   > it is better to get data sent and used by the   
   > receiver than to wait around for more.   
   >   
   > I suggested UDP might be better, to avoid a delay   
   > due to retransmission such that data sent later would   
   > be available to the receiver.   
   >   
   > (That doesn't generally happen for file transfer, though.)   
   >   
   > -- glen   
      
   Maybe we're confused about the chunking.  He said he's gzipping the   
   data, then sending that data in chunks.  I assume he meant that he's   
   breaking the gzip'ed file down into even smaller chunks.  But since gzip   
   is a stream compression mechanism, I don't think you can decompress it   
   if you process them out of order.  So losing a chunk means the rest of   
   that file is lost.   
      
   --   
   Barry Margolin, barmar@alum.mit.edu   
   Arlington, MA   
   *** PLEASE don't copy me on replies, I'll read them in the group ***   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)