Analysis of the results of AB pressure measuring tools

See a lot of tools about AB pressure measurement, but very few of the results of detailed analysis, today I would like to explain

Environment: Nginx server, Nginx high concurrency, save memory

Results Analysis:

this is apachebench, version 2.3copyright 1996 adam twiss, zeus  technology ltd, http://www.zeustech.net/licensed to the apache software  foundation, http://www.apache.org/benchmarking 192.168.80.157  (be patient) Completed  400 requestscompleted 800 requestscompleted 1200 requestscompleted 1600  requestscompleted 2000 requestscompleted 2400 requestscompleted 2800  requestscompleted 3200 requestscompleted 3600 requestscompleted 4000  requestsfinished 4000 requestsserver software: apache/2.2.15server hostname:  192.168.80.157server port: 80document path: /phpinfo.php# test Page document length:  50797 bytes# Page Size concurrency level: 1000# The number of concurrent tests time taken for tests: 11.846  seconds# the duration of the entire test complete requests: The number of requests completed by  4000# failed requests: 0# The number of failed requests write errors: 0total transferred:  204586997 bytes# The amount of network traffic throughout the process html transferred: 203479961 bytes# the amount of HTML content transferred throughout the process requests  per second: 337.67 [#/sec]  (mean) #最重要的指标之一, equivalent to the number of transactions per second in LR, followed by mean in parentheses to indicate that this is an average time per  request: 2961.449 [ms]  (mean) #最重要的指标之二, equivalent to the average transaction response time in LR, followed by mean in parentheses, which indicates that this is an average time per  request: 2.961 [ms]  (mean, across all concurrent requests) # The average transfer rate: 16866.07 [kbytes/sec] received# per connection request actual run time averages the traffic on the network every second, Can help troubleshoot problems with extended response times due to excessive network traffic connection times  (MS) Min mean[+/-sd] median maxconnect:  0 483 1773.5 11 9052Processing: 2 556 1459.1 255  11763waiting: 1 515 1459.8 220 11756total: 139 1039 2296.6 275  11843# the decomposition of time consumed on the network, the specific algorithm of each data is not very clear percentage of the requests served within a certain time  (MS) 50% 27566%  29875% 32880% 37390% 326095% 907598% 926799% 11713100% 11843  (Longest  request) #整个场景中所有请求的响应情况. Each request in the scene has a response time, where 50% of the user's response time is less than 275 milliseconds, 66% of the user response time is less than 298 milliseconds, and the maximum response time is less than 11843 milliseconds. For concurrent requests, the CPU is not actually processed at the same time, but is rotated on a per-request basis, so basically the first time per request time is approximately equal to the second time per  The request time is multiplied by the number of concurrent requests.

Summary: In the remote Web server stress testing, often the effect is not ideal (because the network delay too large), it is recommended to use the intranet of another or more servers through the intranet to test, so that the data obtained, the accuracy will be much higher. If you only have a single server, you can test it directly locally, which is more accurate than the remote test effect.

This article is from the "it--Brother" blog, please make sure to keep this source http://402753795.blog.51cto.com/10788998/1783386

Analysis of the results of AB pressure measuring tools