System resource monitoring device based on business variation

Abstract

The invention discloses a system resource monitoring device based on business variation, including a centralized configuration module: configuring a list of monitored hosts, collecting monitoring items, collecting time, collecting format, and the corresponding relationship between business modules and SQL statements; a unified collecting module: According to the configuration information of the centralized configuration module, the relevant metadata is collected in real time and sent to the distributed analysis module for real-time analysis; the distributed analysis module: according to the collected metadata statistics traffic indicators and system resource usage, calculate the distribution of system resource consumption , and estimate the resource expansion plan based on the business growth trend; historical data archiving module: compress and archive the metadata, and offline the earlier part of the metadata according to the configured offline time; archive the result data for report display and historical trend analysis . The invention makes resource consumption clear, capacity expansion and optimization requirements predictable, greatly reduces failure rate and management cost, and improves system stability.

Description

technical field [0001] The invention relates to a system resource capacity monitoring device, in particular to a system resource monitoring device based on service variation. Background technique [0002] At present, the data of large and medium-sized enterprises is mainly stored in the database. Through the addition, deletion, modification, query and calculation of data to achieve specific business functions, a certain amount of basic system resources such as CPU, memory or space will be consumed each time to complete the comparison. operation of the database. When the system is relatively idle, there will be no contention between various operations due to sufficient basic resources, but as the business volume increases, the business system will run slowly or even become unavailable due to insufficient basic resources. It is especially obvious during the peak period of business, which seriously affects the stable operation of the business system. [0003] Currently there ...

AI Technical Summary

Problems solved by technology

However, when the scale of hosts exceeds 100 or 1,000, there are dozens of hosts that report alarms every day, too many need to be expanded, and the lack of spare parts requires a long time for unified procurement or it is impossible to expand the capacity itself.

The current method is relatively rough on host resource management, and it is impossible to consider resource expansion as a whole, resulting in remedial actions only when incidents occur. There is no forward-looking control and it also increases management costs.

[0008] Method 2: The user finds that the response is slow and reports a failure, and then the business department responds to the maintenance side for emergency expansion. This phenomenon mainly occurs during peak business hours. The increase in user usage leads to resource usage bottlenecks, but during low business peaks At this time, the phenomenon disappears and the system returns to normal, but because it has already affected the user's use, the user experience has been reduced, resulting in a very passive situation

When a problem occurs, the host that caused the problem is due to resource depletion due to business growth, but the problem cannot be repaired in time due to insufficient spare parts or full expansion

[0009] It can be seen from the above that the above two methods have disadvantages such as high management costs, untimely resource expansion, and difficult assessment of expansion needs. Therefore, it is necessary to improve the existing resource capacity monitoring device

Images