Method and system for coping with live broadcast flow burst

Abstract

The embodiment of the invention provides a method and a system for coping with a live broadcast flow burst. The method comprises the following steps: monitoring and acquiring abnormal frequency of a live broadcast service back end about live broadcast service response in real time; and when a popularity live broadcast watching request initiated by a user is obtained, comparing the abnormal frequency of the live broadcast service back end about the live broadcast service response with a preset first threshold, and forwarding the popularity live broadcast watching request to the live broadcast service back end or the proxy unit according to a comparison result. The method is not limited by capacity expansion time, machine reserve and cost, and maximum availability guarantee is provided; whenabnormal flow bursts, a system at the rear end can be effectively protected, and the avalanche effect is avoided.

Description

technical field [0001] The present invention relates to the field of live broadcast, in particular to a method and system for dealing with bursts of live broadcast traffic. Background technique [0002] At present, live broadcast applications are faced with the problem of service availability guarantee in case of abnormal burst traffic. For example, traffic stars suddenly start broadcasting, Weibo official announcements, hot events, etc. This type of traffic is characterized by high traffic bursts and fast growth. [0003] Existing technologies mostly perform elastic expansion through traffic perception. When abnormal traffic growth is detected or reaches the system expansion threshold, the operation and maintenance system automatically expands to increase the service capacity. [0004] In the process of realizing the present invention, the applicant found that at least the following problems existed in the prior art: [0005] In the existing technology, although traffic s...

AI Technical Summary

Problems solved by technology

[0005] In the existing technology, although traffic sensing is used to automatically expand the capacity and quickly increase the capacity of the service, it is still difficult to guarantee the capacity for the unique traffic characteristics of Weibo live broadcast products - large traffic and fast growth rate

[0006] Limited by time: During the expansion period, the traffic growth is too fast and has exceeded the bottleneck of the system, resulting in reduced system availability

[0007] Restricted by machine reserves for capacity expansion: Insufficient servers required for capacity expansion, resulting in capacity expansion failure

If there is a network delay problem in cross-region expansion, it will affect service performance

Images