home bbs files messages ]

Forums before death by AOL, social media and spammers... "We can't have nice things"

   comp.protocols.tcp-ip      TCP and IP network protocols.      14,669 messages   

[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]

   Message 12,971 of 14,669   
   Pascal Hambourg to All   
   Re: Interpreting Iperf Results   
   27 Jul 09 11:18:21   
   
   d4240285   
   From: boite-a-spam@plouf.fr.eu.org   
      
   Hello,   
      
   newbie123 a écrit :   
   >   
   > 1. The above results are saying that in 0 to 60 sec's a total of   
   > 658MBytes were sent at a rate of 92Mb/sec. Is that true?   
   >   
   > 2. When converting 92 x 60 = 5520 Mbits = 690 Mbyes. It does not match   
   > the 658. Is there something odd or is the result being read incorrect.   
      
   I guess that "MBytes" actually stands for mebibytes instead of   
   megabytes, whereas "Mbits/sec" stands for real megabits/second.   
   mega (M) = 10^6   
   mebi (Mi) = 2^20   
   so 658 MiBytes ~ 690 MBytes.   
      
   > 3. Generating multiple connections by using paralle streams option   
   > does not give the same results. Each stream is hardly going above 18   
   > Mb. Is that normal. What advantage is to have multiple streams to 1   
   > stream.   
      
   Some factors such as latency (round trip time) and window size can limit   
   the actual throughput of a single TCP connection, so multiple TCP   
   connections may help using all the available bandwidth. But on a direct   
   ethernet link, which has a very low latency, there is no advantage.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)   

[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]


(c) 1994,  bbs@darkrealms.ca