A Dynamic Data Flow Scheduling Method for Data Center Based on Hybrid Transmission Mode

Abstract

A dynamic data flow scheduling method for a data center based on a mixed transmission mode. Under a data center network with a mixed transmission mode, the size of the data flow, the network topology diagram formed by the switch through wired connections, and the network topology formed by the switch through wireless connections The interference graph and available wireless connections are used as input, combined with relaxation algorithm, linear programming and other methods to solve the data flow scheduling problem, and finally obtain the data flow transmission path and link usage status as the output of the algorithm. Reasonable scheduling of data streams eliminates hot spots in the data center, solves congestion, and improves the overall performance of the data center. At the same time, the consumption of the solution itself is controlled within a certain range without introducing new burdens to the data center. The present invention can be used in On the premise of ensuring data center consumption, reasonably schedule large data flows in the network, eliminate hotspots and congestion in the data center network, and improve data center performance.

Description

technical field [0001] The invention belongs to the technical field of the Internet, and in particular relates to a dynamic data flow scheduling method of a data center based on a mixed transmission mode. Background technique [0002] In recent years, Internet technology has developed rapidly, and cloud computing technology has been greatly promoted and improved. As the infrastructure of cloud computing, the development of data centers has brought unprecedented benefits to the promotion and popularization of cloud computing. At present, application services based on cloud computing are showing rapid growth and are applied in various fields of life, bringing great convenience to people's work, study, and entertainment. However, with the development of data centers and the growth of large-scale services based on cloud computing technology, the data streams generated by these services show polarization characteristics, including small data streams sensitive to delay and sensit...

AI Technical Summary

Problems solved by technology

However, with the development of data centers and the growth of large-scale services based on cloud computing technology, the data streams generated by these services show polarization characteristics, including small data streams sensitive to delay and sensitive to output. The routing and scheduling of these data streams pose a huge challenge to the data center network

How to efficiently handle mixed data flows has become a bottleneck restricting the development of data center networks

[0003] The combination of delay-sensitive small data flows and output-sensitive large data flows dynamically generates hot connections in the data center network, causing network congestion and severely affecting data center performance

Images