Repository logo
 

Towards emulation of large-scale IP networks using end-to-end packet delay characteristics

Abstract

Network emulation combines concepts from network simulation and measurements and provides a emulated network testbed over which application and protocols can be tested. Existing network emulators are not scalable due to the limitations of available computer hardware infrastructure and the reliance on one-to-one packet mapping and modeling scheme. This research proposes a measurement-based modeling methodology for the design of a network-in-a-box emulator. Methodology aims at overcoming the limitation of computational overhead and end-to-end network system characterization. A framework for large scale IP network emulation, named Overall Trend Replicating Network Emulator Tool (OTRENET), is presented. OTRENET intercepts data packet streams and modify them, based on network system models, in real-time. The complexity and overhead of packet-by-packet mapping and modeling, while producing results consistent with measurements is achieved by a traffic sampling algorithm. Such algorithm monitors traffic metrics in a per-packet level, to dynamically separate it into frames. A comprehensive study of end-to-end packet delay dynamics, in the context of network system modeling, is presented. Theoretical basis, techniques and measurements for network packet delay dynamics characterization and modeling for various sending rate conditions and network stages have been developed. Goodness-of-fit results demonstrate the modeling accuracy for both packet delay and IPG processes for cases where sending bit rate is relatively small compared to the link capacity. However, as the sending bit rate increases, as a fraction of the bandwidth, IPG becomes a better alternative for network system modeling. A novel approach for online modeling end-to-end packet delay dynamics is proposed to address non-stationarity of network systems. Proposed methodology models and captures the network system characteristics taking into account the non-stationarity of the packet delay samples. In general, results presented show that analyzing packet delay processes by modeling the segmented traces yield a better understanding of the network system dynamics.

Description

Rights Access

Subject

IP emulator
network modeling
packet autocorrelation
packet delay
electrical engineering
computer science

Citation

Associated Publications