Typical application scenarios of memory databases

In recent years, the memory database (IMDB) technology has developed rapidly. In addition to inherent high performance, IMDB itself is increasingly evolving towards a fully functional independent DB. The following briefly describes several common IMDB application scenarios, hoping to inspire colleagues who are interested in IMDB technology-1. The largest scale of telecom-charged IMDB applications are concentrated in the telecom field, especially

In recent years, the memory database (IMDB) technology has developed rapidly. In addition to inherent high performance, IMDB itself is increasingly evolving towards a fully functional independent DB. The following briefly describes several common IMDB application scenarios, hoping to inspire colleagues who are interested in IMDB technology-1. The largest scale of telecom-charged IMDB applications are concentrated in the telecom field, especially

In recent years, the memory database (IMDB) technology has developed rapidly. In addition to inherent high performance, IMDB itself is increasingly evolving towards a fully functional independent DB.

The following briefly describes several common IMDB application scenarios and hopes to inspire colleagues interested in IMDB technology --

1. China Telecom Billing
The largest application of IMDB is concentrated in the telecom field, especially in the billing system. Of course, in recent years, we have begun to expand to new telecom business areas, such as core networks, CRM, and Precision Marketing. It is the architecture diagram of IMDB in telecom billing for your reference.

2. Online Securities Transactions

Currently, most securities transactions use J2EE to cache all transaction objects, which will inevitably lead to a large amount of system overhead in the middle layer and increase system and labor costs. Using IMDB to logically classify transactions and cache them on the application server can greatly improve the uniformity of system performance and object access.

3. Regional data collection centers
For large business systems across regions, the existence of regional data centers greatly improves data accuracy, speed, and security. Typical application scenarios include: expressway charging system, Catering Chain billing/revenue system, and Agent ticketing system. The typical system architecture is as follows:


This type of system is characterized by a low-load dedicated machine and a relatively low-cost hardware configuration, however, regional data centers require high hardware configurations to handle highly concurrent tasks that may occur (for example, the restaurant performs a unified check at every day, and the agent ticketing system submits a report on the current month at the end of the month ). Considering that high concurrency does not happen all the time, use the configuration of memory database + lightweight database server to replace the high-performance database server as the regional database server, you can manage and maintain hot data in high concurrency at the lowest cost.
In actual deployment, the memory database can be deployed on both the dedicated terminal machine and regional database server. The dedicated machine can be synchronized to the regional database server in synchronous/asynchronous mode. Finally, the Headquarters data center directly reads the summary data from the memory databases of the regional data centers and processes it accordingly.

4. BI SYSTEM
The BI system is composed of data warehouses and a large number of OLAP applications. The bottleneck of traditional BI systems often comes from database servers, which is exactly the original intention of IBM, Oracle and other companies to actively launch the database all-in-one machine. On the other hand, in the domestic BI market, many customers rushed to the Data Warehouse and BI project, but did not realize the serious system bottlenecks that may arise in the future as the data scale expands. These bottlenecks directly affect the efficiency of report generation. The database all-in-one machine seems to be able to solve customer problems, but IT also comes with high equipment prices, and directly affects the credibility of the customer's IT department throughout the company. With no additional hardware investment, the memory database is used and the existing system is transformed to a certain extent. This problem can be solved to a large extent:

Based on the existing system, the memory database is deployed on the server of the application/BI system, or an additional layer of Data acceleration layer is configured with sufficient funds. It is worth noting that such a deployment method should be very familiar with the workflow of the application/BI system, and the data routing method should be divided manually according to specific logic rules, this allows each application system to run their own OLAP resources in a dedicated data set. If you can smoothly sort out and deploy your business, this architecture not only solves the bottleneck of the existing system, but also can clearly sort out the business process and facilitate system expansion in the future.