A billing system based on cloud computing and its implementation method

Abstract

The invention discloses a billing system based on cloud computing and an implementation method thereof. The billing system uses a distribution server with cloud characteristics and a load balancer. Under the control of the load balancer, the real-time status information of each host machine in the cluster system is sent to different distribution servers, and then sent to the billing server for processing. Billing results, so this billing system can dynamically apply to join distribution server resources when the system is busy or when resources are tight, and automatically release resources when idle, realizing on-demand allocation and on-demand access. This dynamic elastic expansion system can easily meet The demand for large-scale growth of users and services in the future.

Description

technical field [0001] The present invention relates to the technical field of cloud computing and cloud computing operation system, in particular to a cloud computing-based billing system and its implementation method. Background technique [0002] With the large-scale cutover of intelligent network users to OCS (Online Charging System, Online Charging System), the scale of users carried by the OCS system will continue to increase. At the same time, with the rapid development of 3G services, the number of 3G users The rapid growth has brought about a sharp increase in data traffic access, and a large amount of data traffic access has brought great pressure to the load of the OCS system. [0003] The existing technology generally adopts the expansion of the OCS system to improve the system carrying capacity, which has the problem of high cost; or uses the minicomputer-based HA (High Available, high availability cluster), but this method has the following disadvantages: [0...

AI Technical Summary

Problems solved by technology

[0003] The existing technology generally adopts the expansion of the OCS system to improve the system carrying capacity, which has the problem of high cost; or uses the minicomputer-based HA (High Available, high availability cluster), but this method has the following disadvantages:

In the master-slave mode, only the master server undertakes the service, and the slave server does not undertake the service, and the resource utilization rate is low; in the symmetrical mode, the two servers have a master-slave relationship with each other, and when one machine goes down, the other must load all the services of the two servers , poor stability; the cluster mode uses three or more servers to work at the same time, one of which serves as the slave server of the other two or more, which has improved in all aspects compared with the above two modes, but there are also idle resources;

[0005] 2. The stability is not high

Once the minicomputer-based HA mode goes down, it needs a second-level or even minute-level switching time window, which may cause performance shocks and stability problems to the running production system;

[0006] 3. Does not have smooth expansion and upgrade capabilities

Images