1 / 3

EE122 Project

EE122 Project. High Level Description. Goal. To use an interactive network simulation environment that allows students to how end-to-end and network control mechanisms affect various measures of latency and loss How do AIMD protocols perform? Compare various marking-based feedback schemes

mircea
Download Presentation

EE122 Project

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. EE122 Project High Level Description

  2. Goal • To use an interactive network simulation environment that allows students to how end-to-end and network control mechanisms affect various measures of latency and loss • How do AIMD protocols perform? • Compare various marking-based feedback schemes • Quantify the role of routing • Quantify the effects of WFQ schedulers • Understand how delay due to admission control can be traded off against network delay

  3. Approach: Sequence of steps that systematically explores the design space • Assume FCFS scheduling, no packet marking but TCP runs on the end hosts • Does TCP behave as advertised? • What is the relationship between network utilization and delay • Change the routes – how does performance change? • Implement a few different kinds of marking schemes (Jean to fill in here) • What is the tangible effect of each scheme (must measure) • What is the complexity of each scheme? (from code) • Implement priority based virtual queues • Assume no marking, but implement WFQ/PGPS at the nodes. Set equal weights. • Add a marking scheme to the scheduling • Assume that the sources are applications which have different rates. Change the WFQ weights to be proportional to these rates. • The flow control scheme may delay packets before admitting them. Assume some Poisson and some CBR sources. When this delay is added to the network delay, how do 1-5 perform? Does anything change?

More Related