The Community for Technology Leaders
2015 ACM/IEEE Symposium on Architectures for Networking and Communications Systems (ANCS) (2015)
Oakland, CA, USA
May 7, 2015 to May 8, 2015
ISBN: 978-1-4673-6633-5
pp: 29-38
Sebastian Gallenmuller , Dept. of Inf., Tech. Univ. Munchen, Garching, Germany
Paul Emmerich , Dept. of Inf., Tech. Univ. Munchen, Garching, Germany
Florian Wohlfart , Dept. of Inf., Tech. Univ. Munchen, Garching, Germany
Daniel Raumer , Dept. of Inf., Tech. Univ. Munchen, Garching, Germany
Georg Carle , Dept. of Inf., Tech. Univ. Munchen, Garching, Germany
ABSTRACT
Network stacks currently implemented in operating systems can no longer cope with the packet rates offered by 10 Gbit Ethernet. Thus, frameworks were developed claiming to offer a faster alternative for this demand. These frameworks enable arbitrary packet processing systems to be built from commodity hardware handling a traffic rate of several 10 Gbit interfaces, entering a domain previously only available to custom-built hardware. In this paper, we survey various frameworks for high-performance packet IO. We analyze the performance of the most prominent frameworks based on representative measurements in packet forwarding scenarios. Therefore, we quantify the effects of caching and look at the tradeoff between throughput and latency. Moreover, we introduce a model to estimate and assess the performance of these packet processing frameworks.
INDEX TERMS
Software, Hardware, Throughput, Linux, Upper bound, Bandwidth, Multicore processing
CITATION

S. Gallenmuller, P. Emmerich, F. Wohlfart, D. Raumer and G. Carle, "Comparison of frameworks for high-performance packet IO," 2015 ACM/IEEE Symposium on Architectures for Networking and Communications Systems (ANCS), Oakland, CA, USA, 2015, pp. 29-38.
doi:10.1109/ANCS.2015.7110118
196 ms
(Ver 3.3 (11022016))