In mobile computing, various energy-efficient thread scheduling schemes for heterogeneous multiprocessing architectures are proposed. For network applications, however, inaccurate prediction of the CPU load and high-priority network packet processing overdrive CPU cores, leading to large energy consumption. We present a framework including a network stack monitor, a bandwidth controller, and an energy-efficient thread scheduling scheme, which accurately estimates the CPU load for packet processing and optimally schedules CPU resource. It improves performance/watt by 4.79 times over the baseline Linux scheduler for FTP applications. In multi-threaded environments, it improves performance/watt by 2.35–3.11 times for PARSEC benchmark and network applications.

This paper provides insights into the traffic flow monitoring system of OF@TEIN (OpenFlow@Trans Eurasia Information Network) testbed. OF@TEIN is software defined networking (SDN) testbed adapted by KOREN (KOrea advanced REsearch Network) and integrated with the research and education networks of several Asian nations including Japan, Malaysia, Thailand, Vietnam, Philippine, etc. Traditional traffic monitoring solutions such as NetFlow, RSPAN ports, Network Packet Brokers (NPBs) can't provide network-wide visibility in OF@TEIN because OF@TEIN is a large multi-tenant testbed deployed over high speed research networks across several countries. Therefore, we have implemented a new sFlow-based flow monitoring system that is tailored to OF@TEIN requirements and can provide real-time L2 to L7 network wide visibility. OF@TEIN uses SDN-based network virtualization to slice the network among multiple concurrent experimenters. Machines in a network slice (or VLAN) communicate with each other using GRE tunnels. Our traffic monitoring system enables monitoring flow spaces of VLANs as well as physical provider network. It utilizes northbound interfaces (NBIs) exposed by sFlow-RT analytics engine and FloodLight SDN Controller. It periodically fetches flows statistics from sFlow-RT and stores them in time-series format in Whisper RRD database. Graphite real-time charting tool is used to plot the statistics stored in Whisper.