[e2e] Re: [Tsvwg] What's the benefit of out-of-order processing?

[Phillip T. Conrad]

In response to the question 

>Is there any study done on evaluating the effect of this TCP
>"deficiency"?  What applications really need to and are capable to do
>out-of-order processing? 

I would call your attention to the following journal articles and conferences papers:

        Computer Networks, 31(7), 4/99, P. Amer, S. Iren, G. Sezen, P. Conrad, M. Taube, A. Caro 
        "Network-conscious GIF image transmission over Internet"

        ACM Multimedia 2001 (to appear, 9/01) P. Conrad, A. Caro, P. Amer,  
        "ReMDoR: Remote Multimedia Document Retrieval over Parital Order Transport"

        IEEE MILCOM 2001 (to appear, 10/01) P. Conrad, G. Heinz, A. Caro,  P. Amer, J. Fiore, 
        "SCTP in Battlefield Networks"

These, and over a dozen more papers and articles on the benefits of out-of-sequence processing, (including three Ph.D. Dissertations addressing the topic) can be found at http://www.eecis.udel.edu/~amer/PEL/poc/index.html.

I would also call attention to the seminal paper in this area:

        Clark and Tennenhouse, 1990: D. Clark and D. Tennenhouse. Architectural 
        Considerations for a New Generation of Protocols. Proc. SIGCOMM ‘90.

as well as to the work of Diot and Gagnon, (which while I argue in my dissertation---see the link above---that it understates the benefit of out-of-sequence processing, is nevertheless useful in establishing some baselines for the benefit.)

        Diot and Gagnon, 1999: C. Diot, F. Gagnon. Impact of out-of-sequence processing
        on the performance of data transmission. Computer Networks, 31(5), 475-492,
        Mar. 1999.

I hope these are helpful.

Phill Conrad
Asst. Professor, CIS Dept. Temple University.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.postel.org/pipermail/end2end-interest/attachments/20010918/1edbe7d3/attachment.html