A method and device for precise cluster speed limiting

Abstract

Embodiments of the present application provide a cluster precise rate limiting method and device, which relate to the technical field of computers. The method of the present application includes: receiving a data packet; judging whether the data packet is a data packet of a client that is responsible for handling by itself; if the data packet is not a data packet of a client that is responsible for handling by itself, forwarding the data packet To the service entity responsible for dealing with the client to which the data packet belongs; if the data packet is a data packet of the client that it is responsible for handling, then based on the identity information of the data packet, perform rate-limiting processing on the data packet. This application does not need to separately set a flow control entity in the cluster to limit the speed of the service entities in the entire cluster, and the failure of the flow control entity will not affect the normal operation of the entire cluster, which has strong applicability. Moreover, it is only simple forwarding of data packets between service entities, and the complexity is low.

Description

technical field [0001] The present application relates to the field of computer technology, in particular to a cluster precise speed limiting method and a cluster precise speed limiting device. Background technique [0002] With the popularization of the network, the application environment of the network becomes more and more diversified, for example, a cluster composed of a group of service entities working together is used to provide services to multiple clients or multiple users. In such an application environment, in order to prevent the excessive traffic of a certain client from affecting other clients or based on the demand for broadband traffic purchased by a certain client, it is necessary to limit the traffic rate of each client or user. Since the cluster contains multiple service entities, such as servers, etc. Each service entity provides services at the same time, which will cause the traffic of the same client to fall on multiple service entities in the cluste...

AI Technical Summary

Problems solved by technology

[0008] For the first type of solution, it is first necessary to provide a flow control entity in the cluster to control the rate limit. If the flow control entity fails, it will affect the normal operation of the cluster, and the applicability is not strong

Secondly, because the flow control entity needs to monitor the traffic rate of the service entity, it needs to communicate with the service entity in a large amount to decide the rate limit issued to each service entity, which is complex

[0009] For the second type of solution, it is difficult to ensure that the traffic of each client falls evenly on each service entity, and the operability is low

Secondly, if the traffic of the client is uneven, there will be traffic on some service entities for some clients that is greater than the specific speed limit value, resulting in packet loss, causing the actual bandwidth of the user to be less than the bandwidth purchased by the user, and the accuracy is relatively low. Low

[0010] For the third type of solution, adding a dedicated speed limit node will increase the cost, which is not applicable to small clusters

Images