Remember that TCP only understands a byte stream, not the structure
of application records within it. Brian is correct about one
application for interactive traffic.
Another might be if you had, let's say, 2000 byte application
records, but the window size had worked its way up to 64 K. You
might not want to wait for the window to fill before a given record
was sent, so you might push it. Doing so does not optimize
throughput, but would improve application latency. Everything is a
tradeoff.
>From Comer's V1 3rd edition..
>
>Looks like a way to force data transmission, 1 quoted use is after each
>keystroke of an interacive application.
>
> Brian
>
>----- Original Message -----
>From:
>To:
>Sent: Friday, June 15, 2001 12:38 PM
>Subject: PSH: Push Function in RFC 793 [7:8756]
>
>
>> Could someone give me a brief decription of the PSH: Push Function in
>TCP?
>>
>> Control Bits: 6 bits (from left to right):
>>
>> URG: Urgent Pointer field significant
>> ACK: Acknowledgment field significant
>> PSH: Push Function
>> RST: Reset the connection
>> SYN: Synchronize sequence numbers
> > FIN: No more data from sender
Message Posted at:
http://www.groupstudy.com/form/read.php?f=7&i=8890&t=8756
--------------------------------------------------
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
