Traffic Generation for Benchmarking Data Centre Networks

Published in Optical Switching and Networking, 2022

Recommended citation: C. W. F. Parsonson, J. L. Benjamin and G. Zervas, "Traffic Generation for Benchmarking Data Centre Networks", Optical Switching and Networking, 2022 https://www.sciencedirect.com/science/article/pii/S1573427722000315

Benchmarking is commonly used in research fields, such as computer architecture design and machine learning, as a powerful paradigm for rigorously assessing, comparing, and developing novel technologies. However, the data centre network (DCN) community lacks a standard open-access and reproducible traffic generation framework for benchmark workload generation. Driving factors behind this include the proprietary nature of traffic traces, the limited detail and quantity of open-access network-level data sets, the high cost of real world experimentation, and the poor reproducibility and fidelity of synthetically generated traffic. This is curtailing the community's understanding of existing systems and hindering the ability with which novel technologies, such as optical DCNs, can be developed, compared, and tested. We present TrafPy; an open-access framework for generating both realistic and custom DCN traffic traces. TrafPy is compatible with any simulation, emulation, or experimentation environment, and can be used for standardised benchmarking and for investigating the properties and limitations of network systems such as schedulers, switches, routers, and resource managers. We give an overview of the TrafPy traffic generation framework, and provide a brief demonstration of its efficacy through an investigation into the sensitivity of some canonical scheduling algorithms to varying traffic trace characteristics in the context of optical DCNs. TrafPy is open-sourced via GitHub and all data associated with this manuscript via RDR.

View paper here