A business processing method, an Internet of Things billing infrastructure system, and a storage medium

Abstract

The embodiment of the present invention discloses a business processing method. According to the preset classification and grading rules, the original data is classified and stored according to the classification and classification, and the preset database is obtained; the original computing resources are classified according to the central processing unit resources and the graphics processing unit resources through the master node. , and central processor and graphics processor resources are packaged and divided to obtain preset computing resources; through the master node, according to the corresponding relationship between preset business and scheduling content, determine the target database corresponding to the target business in the preset database, and the preset Design the target computing resources corresponding to the target business in the computing resources; call the target database and target computing resources through the master node to execute the target business.

Description

technical field [0001] The present invention relates to the technical field of data processing, in particular to a service processing method, an Internet of Things charging infrastructure system and a storage medium. Background technique [0002] With the rapid development of the Internet of Things, its user scale and business volume continue to increase, which poses a huge challenge to the core billing system supported by the Internet of Things. [0003] It is easy to consume a lot of time when using the existing infrastructure for billing, and even many services with Internet of Things characteristics cannot be launched at all. For example, currently a distributed database can store millions of users under one account, but the performance of account withdrawals for large accounts is limited by the accumulative account fees. hours or so. Or, because it takes too long to accumulate billing expenses in real time, the accounts of millions of users do not support the query of...

AI Technical Summary

Problems solved by technology

[0003] It is easy to consume a lot of time when using the existing infrastructure for billing, and even many businesses with Internet of Things characteristics cannot be launched at all

For example, currently a distributed database can store millions of users under one account, but the performance of account withdrawals for large accounts is limited by the accumulative account fees. about hours

Or, because the real-time accumulated bill expenses take too long, the accounts of millions of users do not support the query of unbilled expenses

[0004] The reason for the above problems is that the existing infrastructure system has low computing power in the face of massive connections and high concurrent data

Therefore, many enterprises begin to use the central processing unit (Central Processing Unit, CPU) + graphics processing unit (Graphics Processing Unit, GPU) asynchronous architecture to solve the problem of computing efficiency of massive calculations, but in the prior art, there are two problems in this method The core division of labor is not clear, the core communication efficiency is low and other defects, and the processing efficiency of the billing system is low

Images