System monitoring data collecting method with dynamically controllable traffic
A technology for monitoring data and traffic dynamics, applied in hardware monitoring and other directions, can solve problems such as the overall performance and stability of the system, and achieve the effect of ensuring quality
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0033] figure 1 It is the monitoring data asynchronous acquisition and processing structure diagram of the present invention. The working principle of this part is that by calling the Linux system thread interface, the monitoring module starts a separate monitoring data processing thread (ProcessThread) after initialization to be responsible for the multi-target processing output of system data , each time the main program calls the data acquisition interface, it actually only caches the system state data internally to a data output request queue in memory, and does not generate actual IO operations; then, after the acquisition interface completes the queue operation, it will Return immediately and let the main program continue to execute its real-time application logic, thus essentially avoiding the danger of the IO that may be caused by the synchronous operation to hang the main program.
[0034] The monitoring data processing thread (ProcessThread) continuously polls the da...
Embodiment 2
[0037] The collection interface of the monitoring data collection library will check the occurrence frequency of system monitoring events in real time and dynamically adjust the flow of collected data when performing data enqueuing operations, so as to avoid the impact of a large amount of data IO operations on the overall performance of the system. This part is designed as follows: the system sets three parameters, namely, the flow control detection cycle (rateControlInterval), the data output request number threshold (rateControlCount) and the flow control execution cycle (rateControlDuration). The flow control algorithm first counts the number of data requests arriving in the queue during the flow control detection period. Once the number of requests reaches or exceeds the request threshold, the flow control mechanism is immediately activated and lasts until the end of the flow control execution period. Therefore, a total cycle of dynamic flow control can be divided into two...
Embodiment 3
[0039] In the flow control execution cycle, the collection interface of the monitoring data collection library no longer adds the newly arrived data to the data output request queue, but merges the same characteristic data through the pre-established hash table during initialization, so as to reduce the actual data. traffic target. You need to define the concept of "same characteristic data" (that is, Hash Key) according to your own business. For example, for a securities trading system, if the monitoring data comes from the same PBU or product number, then it can be defined as the same characteristic data. The data merging mechanism is: for a feature data type that has not appeared before, add an item in the e hash table; otherwise, only increase the count of the same record type.
[0040] Take the monitoring of system errors as an example: when a large number of errors burst out, the acquisition interface writes each piece of error data with the same characteristic data of t...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com