Packet Network Simulation: Speedup and Accuracy Versus Timing Granularity

Ahn, Jong Suk
Danzig, Peter B

Abstract

This paper describes a new technique that can speedup sirmdatioo of high-speed, wide-area packet networks by one to two order of magnitude Speedup is achieved by coarsening the representation of network traffic from packet-by-packet to train-by-train, where a train represents a cluster of closely spaced packets. Coamening the timing granularity creates longer trains and makea the simulation pmeeed more quickly since the coat of proeeaaing trains is independent of train size. Coarsening the timing granularity introduces, of course, a degree of approximation. This paper presents experiments that evaluate our coarse time-grain simulation technique for first idtlrst out (FIFO) switched, intemet protocol (W), and asynchronous transfer mode (ATM) networks carrying a mix of data and streaming traftic. We show that delay, throughput, and loss rate ean frequently be estimated within a few percent via coarse time-grain simu-lation. Thii paper also deseribes how to apply coarse time-grain simulation to other switch dia&plines. Finally, this paper introdssees three more simulation techniques which together can double the performance of well written packet simulators without trading w-itb the shmdation accuracy. These teehniquea reduce the number of outstanding simulation events and reduce the cost of manipulating the event list.

Keywords

calendar
flowsim
speedup
hybrid
packet train

Notes

Introduces Jain's idea of packet trains. Discusses flowsim, lookahead and some hybrid ideas.

Related Papers

Bibtex

 @article{ ahn.danzig_speedup96,
    author = "J. Ahn and P. Danzig",
    title = {Packet network simulation: Speedup and accuracy versus timing granularity},
    journal = "IEEE/ACM Transactions on Networking",
    volume  = "4",
    number = "5",
    month  = "Oct.",
    pages = "743 -- 757",
    year = "1996"
}

Back to Intro By Author By Importance By Keyword By Title By Reference