Method for optimizing throughput rate and response time of decision engine

Abstract

The invention provides a method for optimizing the throughput rate and response time of a decision engine. The method comprises the following steps that: 1, the task buffer pool of tasks are established by using a blocking queue, the tasks are initialized, the tasks are putted into the queue, the task at the first place is taken out, a value is assigned to the current task, and the initializationis finished; 2, when a request arrives, the request is processed by using the current task, and the current task is logged out when a counter is requested and the modulo of a threshold value is 0; 3,a new task is generated and inputted into the queue; and 4, the logged-out task and the generated task are merged into the queue. According to the method, the task buffer pool is constructed, JVM garbage collection operation caused by the fact that a large amount of garbage is generated when the request executes task deletion each time can be avoided; batch centralized destruction is carried out through modulus operation, so that the response time instability of a rule engine caused by garbage collection under high concurrency of a system is avoided.

Description

technical field [0001] The invention mainly relates to the field of decision engines, in particular to a method for optimizing the throughput rate and response time of the decision engine. Background technique [0002] The decision engine originated from the rule-based expert system and developed from the reasoning engine. It is an independent module or component widely used in the Internet industry. It realizes the separation of business decisions from application code and uses predefined semantics Modules write business decisions, accept data input, interpret business rules, and make business decisions based on business rules. The decision engine mainly includes three parts, the rule base, the fact base (working memory) and the reasoning engine. The functional positioning of the decision engine middleware determines that the decision engine shoulders important tasks. Most of these work decision engine frameworks have been implemented, but when specific business scenarios ...

AI Technical Summary

Problems solved by technology

[0003] In actual business scenarios, high concurrency will frequently put real objects into tasks and destroy real objects through the API. Such operations will bring frequent JVM GC garbage collection, thereby reducing system response efficiency, and system TPS and response time will decrease. Frequent fluctuations, this is because the JVM GC will use CPU resources, resulting in a short-term interruption of the system at the millisecond level, thus affecting the TPS and response time of the service

Images