Energy conservation method based on combination of data center network routing and flow preemptive scheduling

Abstract

The invention provides an energy conservation method based on combination of data center network routing and flow preemptive scheduling. The energy conservation method comprises the following steps: network flow is received, and multiple routing paths of the network flow transmitted in a data center network are calculated; whether an idle path exists in the multiple routing paths is judged; when no idle path exists in the multiple routing paths, an alternative routing path set is obtained from the multiple routing paths according to the priority level of the network flow; first routing paths are selected from the alternative routing path set, wherein the first routing paths which pass through an idle switchboard is the smallest in number, and the network flow is sent through the first routing paths. By means of the energy conservation method based on the combination of the data center network routing and the flow preemptive scheduling, the utilization rate of network resources can be improved effectively, the energy using efficiency of the data center network is improved substantially, meanwhile, average completion time of the network flow can be improved, real-time performance and feasibility of data transmission are ensured, and the data transmission is made to be more flexible and convenient to control.

Description

technical field [0001] The invention relates to the field of network technology, in particular to an energy-saving method based on the combination of data center network routing and flow preemption scheduling. Background technique [0002] The goal of the data center network is to interconnect a large number of data center servers to provide effective and fault-tolerant routing and transmission services for upper-layer applications. As the infrastructure of cloud computing, the construction of data center network needs to meet the goals of high scalability, high reliability, high network throughput and low cost. However, with the rise of cloud computing applications and the continuous expansion of the scale of data centers, the problem of energy consumption in data centers has become prominent. The huge energy consumption of data centers not only leads to huge network operating costs, but also has serious negative impacts on society and the environment. It has become a majo...

AI Technical Summary

Problems solved by technology

However, due to competition among network flows for bandwidth, in these energy-saving routing schemes, network flows cannot fully utilize network bandwidth resources, resulting in low network link utilization and low efficiency of network power usage.

In addition, the existing flow scheduling strategies usually focus on how to achieve load balancing during network flow transmission to obtain high-performance network flow transmission, but do not use flow scheduling strategies to achieve network energy saving

Images