Shulou(Shulou.com)06/02 Report--

This article introduces the relevant knowledge of "what are the reasons for not going to TPS in the performance test". In the operation of actual cases, many people will encounter such a dilemma, so let the editor lead you to learn how to deal with these situations. I hope you can read it carefully and be able to achieve something!

1. Network bandwidth

In stress testing, it is sometimes necessary to simulate a large number of user requests. If the data packets transmitted per unit time are too large and exceed the transmission capacity of bandwidth, it will cause competition for network resources. Indirectly, the number of requests received by the server can not reach the upper limit of the processing capacity of the server.

2. Connection pool

The number of available connections is too small, causing the request to wait. Connection pooling is generally divided into server connection pooling (such as Tomcat) and database connection pooling (or the maximum number of connections allowed).

(for details on connection pooling, please refer to the previous blog: performance testing: connection pooling and threads)

3. Garbage collection mechanism

For common application servers, such as Tomcat, because the stack memory of java is allocated dynamically, and the specific recovery mechanism is based on algorithms, if the new generation of Eden and Survivor areas are Minor GC frequently, and the old full GC is also recycled more frequently, then TPS

It also has a certain impact, because garbage collection itself will occupy a certain amount of resources.

4. Database configuration

In the case of high concurrency, if the request data needs to be written to the database and multiple tables need to be written, if the maximum number of connections to the database is not enough, or if the SQL that writes the data has no index, no bound variables, or no master-slave separation, read-write separation, etc.

It will cause database transactions to be processed too slowly, affecting TPS.

5. Communication connection mechanism

Serial, parallel, long connection, pipe connection and so on, different connection conditions will also indirectly affect the TPS.

(for the connection of the protocol, please refer to the previous blog: HTTP Protocol Advanced: connection Management)

6. Hardware resources

Includes CPU (configuration, utilization, etc.), memory (occupancy, etc.), disk (Imax O, page swapping, etc.).

7. Press

For example, for jmeter, the load capacity of a single machine is limited, and if the number of user requests that need to be simulated exceeds its load limit, it will indirectly affect TPS (at this time, distributed pressure testing is needed to solve the problem of single machine load).

8. Stress test script

Take jemter as an example, as encountered by colleagues in previous work, the maximum number of simulation requests exceeds the set number of threads, resulting in a shortage of threads.

When it comes to this reason, it means that sometimes reasons such as the parameter configuration of the test script will also affect the test results.

9. Business logic

The service decoupling degree is low and complex, and the problem caused by the elongation of the whole transaction line.

10. System architecture

For example, whether there is a cache service, cache server configuration, cache hit ratio, cache penetration and cache expiration will affect the test results.

This is the end of the content of "what are the reasons for not going to TPS in the performance test". Thank you for your reading. If you want to know more about the industry, you can follow the website, the editor will output more high-quality practical articles for you!

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.