API performance optimization experiment device

Abstract

The invention discloses an API (Application Program Interface) performance optimization experiment device, which comprises an experiment management module, an experiment deployment module which is used for packaging and mirroring the optimized branch codes corresponding to the experiment names and distributing starting containers; and a service registration discovery module which is used for routing a corresponding request through the service name and the experiment name. The beneficial effects are that multiple performance experiments can be carried out at the same time, and the engineering period is shortened; the multi-layer performance experiments are mutually independent, so that the accuracy of data statistics is not influenced; each layer of experiment has 100% flow, so that enoughexperiment data can be guaranteed; rapidly creating an experiment; each experimental group deployment environment of each layer is physically isolated, so that the influence of other external factorson experimental results is reduced; second-level loss stopping is monitored in real time in an experiment.

Description

technical field [0001] The invention relates to the field of software engineering, in particular to an experimental device for optimizing the performance of an API interface. Background technique [0002] Performance has always been an important indicator for measuring software systems, which directly affects the user experience. With the increase in the number of concurrent users that the system can withstand, how to quickly optimize and iterate the existing system from the code level under the existing hardware conditions such as bandwidth, cpu, memory, and hard disk is a huge test. Because it is difficult to simulate the real online environment in offline testing, only online real data can give the most accurate performance feedback, so online experiments become very important. Under the existing technology, if multiple technical optimization schemes are verified and tested in the experiment at the same time, there may be mutual exclusion between each technical scheme, a...

AI Technical Summary

Problems solved by technology

[0003] The existing technology uses a single-layer experiment to test the impact of different technical optimization schemes on performance indicators. Multiple technical optimization schemes need to be launched in multiple phases, and each time a new experiment is created, additional development investment is required. Therefore, developing new functions in the original On the basis of increasing manpower and time costs, lengthening the delivery cycle, delaying the product's response to the market, and reducing the speed of development and optimization iterations

Another solution in the existing technology is to divide the traffic into multiple parts and carry out multiple experiments at the same time, but excessive traffic refinement can easily lead to too little data and affect the reliability of test result evaluation

Images