Re: draft-cameron-tmux-02.txt
Pete Cameron <cameron@xylint.co.uk> Tue, 08 February 1994 18:04 UTC
Received: from ietf.nri.reston.va.us by IETF.CNRI.Reston.VA.US id aa12498; 8 Feb 94 13:04 EST
Received: from CNRI.RESTON.VA.US by IETF.CNRI.Reston.VA.US id aa12494; 8 Feb 94 13:04 EST
Received: from basil.xylint.co.uk by CNRI.Reston.VA.US id aa12874; 8 Feb 94 13:04 EST
Received: by basil.xylint.co.uk (4.1/SMI-4.1) id AA28196; Tue, 8 Feb 94 17:28:33 GMT
Date: Tue, 08 Feb 1994 17:28:33 +0000
Sender: ietf-archive-request@IETF.CNRI.Reston.VA.US
From: Pete Cameron <cameron@xylint.co.uk>
Message-Id: <9402081728.AA28196@basil.xylint.co.uk>
To: dee@skidrow.lkg.dec.com
Subject: Re: draft-cameron-tmux-02.txt
Cc: cmp-id@xylint.co.uk
--------- Received message begins Here --------- > From: "Donald E. Eastlake 3rd (Beast)" <dee@skidrow.lkg.dec.com> > >The consensus of opinion at the Houston BOF was that the > >document should be made very specific to terminal servers, > >hence the direction I am taking. Also, it matters a LOT what > >kind of packets are passed from the higher level to TMux. We > >tried some experiments with ftp to see what would happen. If > >we allow ftp packets to be multiplexed, then for a single > >ftp session we got a 15% degradation in performance. We > >consider this unacceptable for something that is designed to > >increase performance. Having said this, I still agree to a > >degree with your sentiments, TMux is applicable to all IP > >trafic, but must be used with care. > > Well, why was that? You already recommend against tmux'ing large > packets, which is reasonable. Was it the delays in ACKs or something? Yes, we believe that is the case. > Even if this caused degradation for a single ftp session, would it have > increased throughput if there had been 10 ftp sesssions? It actually improved performance for 2 or more sessions to and from a pair of hosts. > It would be excellent if you could characterize what benefits most > from tmux other than saying "terminal servers"... My characterisation of the benefits, are on those protocols where an increased delay is acceptable. In my view, these are the protocols where people are the recipients of the data. People can not notice short delays <50ms and accept short delays <200ms. That is why we talk about login sessions (changed to interactive sessions in next draft). Pete
- Re: draft-cameron-tmux-02.txt Donald E. Eastlake 3rd (Beast)
- draft-cameron-tmux-02.txt David A. Borman
- Re: draft-cameron-tmux-02.txt Dave Crocker
- Re: draft-cameron-tmux-02.txt Pete Cameron
- Re: draft-cameron-tmux-02.txt Donald E. Eastlake 3rd (Beast)
- Re: draft-cameron-tmux-02.txt Pete Cameron
- Re: draft-cameron-tmux-02.txt Donald E. Eastlake 3rd (Beast)
- Re: draft-cameron-tmux-02.txt Pete Cameron
- Re: draft-cameron-tmux-02.txt Donald E. Eastlake 3rd (Beast)
- Re: draft-cameron-tmux-02.txt Pete Cameron
- Re: draft-cameron-tmux-02.txt Donald E. Eastlake 3rd (Beast)