home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.protocols.tcp-ip
- Path: sparky!uunet!destroyer!fmsrl7!lynx.unm.edu!nmsu.edu!amolitor
- From: amolitor@nmsu.edu (Andrew Molitor)
- Subject: Re: Large Windows and Delayed Acking
- Message-ID: <1993Jan22.162817.3224@nmsu.edu>
- Sender: usenet@nmsu.edu
- Organization: Department of Mathematics
- References: <1993Jan21.233655.29402@gordian.com>
- Date: Fri, 22 Jan 1993 16:28:17 GMT
- Lines: 33
-
- In article <1993Jan21.233655.29402@gordian.com> mike@gordian.com (Michael A. Thomas) writes:
- >
- > What I have found is that with large windows (say 16kb) the sending
- >TCP stops sending at about 4k and waits for the receiving TCP to
- >send an ACK. The receiving TCP, however is using a defered acking
- >scheme which tries to delay sending ACK until a larger window is
- >consolidated into one packet. The metric it is using is 1/2 the
- >buffer size, and the delayed ack period is on the order of 100ms.
- > I can understand why a sending TCP might be leary of sending *so*
- >much data without hearing an ack, but my question is what algorithm
- >is the sending TCP using to determine how much data it will send
- >before going into this quiesent state? Is there an appropriate
- >RFC which I should read which describes this situation? Is the
- >behavior of the sending TCP correct?
-
- Sounds like the sender's computing a congestion window
- around 4K. See, say, Steven's _Internetworking with TCP/IP_ or
- probably rfc793 somewhere or other. Steven's has a good exposition.
-
- The point is that TCP makes a guess at a good sending
- window size based on latency of packets, and uses the minimum
- of that window size and the remote's advertisement as the real
- window. This is used to avoid congestion. It might even be that
- the delayed ACK cause the sender to compute an artificially
- small congestion window (4K seems small to me).
-
- One must keep in mind that TCP is polite. If the goal is
- to hammer your media to death, TCP is the wrong protocol. Not, by
- the way, to suggest that this is your goal.
-
- Andrew
-
- > Michael Thomas (mike@gordian.com)
-