Shulou(Shulou.com)06/03 Report--

Background introduction:

Because "XX Reading platform" is a public welfare project, volunteers participate in product design, development and testing, and the project organization is loose. There is no benchmark for performance analysis and investigation of such products, and we can only work according to common performance analysis strategies. According to the characteristics of the project products, the performance analysis and testing strategies are as follows:

First phase of product

Product goal: to complete 60% of the level 1 and sub-functions. After the client development of the two platforms, only functional demo is used to show the prototype of the product, mainly for the display of articles and books.

Performance testing strategy: no performance testing for the time being; you can prepare the test data.

The second phase of product

Product goal: complete 80% of the function

Target users: monks and mages in XXX monastery

Performance testing goal: complete server performance testing and tuning before launch

Performance testing strategy:

1. Common user interfaces and operations

2. Interfaces with frequent DB interactions

3. Consumptive operation: such as picture loading consumes CPU, book download test bandwidth, maximum number of connections for online reading support

The third phase of product

Product goal: open to volunteers and householders of XXX Temple, and provide application market download and use according to the product situation.

Target user: real user

Performance testing goal: complete client performance testing and tuning before launch

Performance testing strategy:

1. Time consuming of critical operations on the client

two。 Power consumption for continuous use of content over a certain period of time

3. Flow loss for critical operations

4. In order to ensure the fluency of the application, it is necessary to analyze the memory consumption.

5. Since the developers come from all over the world, we hope to establish a flashback log upload function for development and analysis.

Server performance metrics are as follows:

1. The execution of the method takes time in milliseconds

two。 The operation level is time-consuming according to the industry standard 1pm 3pm 5pm 8pm 10. That is, the operation level takes at least less than 3 seconds.

3. Server response time and TPS

4. The stress test observes the server overhead and gives recommendations for the best use of the product:

Base number: CPU 60% pressure test up to 80%

Client performance test scenarios and evaluation criteria, such as screenshot:

1. Screen off (silent) for 1 hour, no message push scenario, power consumption

two。 Memory peak, average

3. CPU share

4. Bright screen operation, CPU

5. First startup, loading time

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.