Statistical Framework for Network Simulation

From Nsnam
Revision as of 18:56, 16 May 2008 by Tjkopena (Talk | contribs)

Jump to: navigation, search

Previous Work

Several components and packages have been made for ns-2 to collect and manage data and statistics. A variety of these are listed in the ns-2 wiki. The following are notes on particular efforts.

  • ns2measure. Main component is a global observer object integrated into ns-2. Several generic types of measures are supported, e.g. time averaged and discrete rate. Observed samples are recorded via an explicit call to the observer object, identified by a measure label and particular identifier such as a flow or host. Post-simulation scripts provide for analyzing collected data, generating statistics. Includes support for incorporating data from independent runs and for conducting runs until a statistical goal such as a confidence is met, but does not provide support for distributing execution of those runs.
  • simd.
  • ns-2/akaroa-2. Provides support for executing and managing from distributed, independent replications. Changes to ns-2 include adding a random number generator with better guarantees of independent streams. Runs a master programs on one computer, and a set of clients on the others that execute received simulations. Within each ns-2 instance a global observer is created. Samples are reported to that observer, which in turn reports them back to the master computer. Measures are identified in simulation scripts by numeric identifiers and consist of particular observations, e.g. delay or packet size. The master program receives these observations and calculates statistics such as the mean and confidence interval over them. That data is used both for final output, and to conduct more simulations at the client machines if confidence is low.
  • tracegraph.
  • rpi ns2graph.
  • ns2 jtrana. Takes an ns-2 trace file, parses it into MySQL, and provides an interface to interrogate the database and produce graphs and other output in several formats. The database scheme seems to be a straightforward encoding of the raw trace data.
  • Samer Bali's tools. Scripts for post-simulation trace analysis, and averaging from multiple runs. Calculates packet delivery ratio, normalized throughput, routing overhead ratio, end to end delay, hop count, connection time.

Raw Notes

fault tolerance