1.worker_processes: worker processes, can be seen by the following command worker_processes default number of worker processes is a worker process
General Configuration number of CPU cores must be configured or arranged directly auto
[root @ localhost ~] # CAT /apps/nginx/conf/nginx.conf|grep Work worker_processes 1; the number of nginx before worker_processes # modify parameters worker process [root @ localhost ~] # PS -ef | grep nginx root 1162 1 ? 0 00:00 00:00:00 nginx: Master Process / Apps / nginx / sbin / nginx -c /apps/nginx/conf/nginx.conf nginx nginx 00:00:00 1225 1162 0 00:19:? worker process the root PTS 01:24 1302 1250 0/0 = 00:00:00 grep --color Auto nginx [the root @ localhost ~] # # the worker process is modified after 2, and a hot start to see the number of worker processes nginx [ ~ @ localhost the root] # VI /apps/nginx/conf/nginx.conf User Nginx Nginx; worker_processes 2; [the root @ localhost ~] # -t Nginx Nginx: Configuration File The syntax /apps/nginx/conf/nginx.conf IS OK nginx: configuration file /apps/nginx/conf/nginx.conf test is successful [root@localhost ~]# nginx -s reload [root@localhost ~]# ps -ef|grep nginx root 1162 1 0 00:00 ? 00:00:00 nginx: master process /apps/nginx/sbin/nginx -c /apps/nginx/conf/nginx.conf nginx 1306 1162 0 01:26 ? 00:00:00 nginx: worker process nginx 1307 1162 0 01:26 ? 00:00:00 nginx: worker process root 1309 1250 0 01:26 pts/0 00:00:00 grep --color=auto nginx [root@localhost ~]# #配置成auto [root@localhost ~]# ps -ef|grep nginx root 1162 1 0 00:00 ? 00:00:00 nginx: master process /apps/nginx/sbin/nginx -c /apps/nginx/conf/nginx.conf nginx 1306 1162 0 01:26 ? 00:00:00 nginx: worker process nginx 1307 1162 0 01:26 ? 00:00:00 nginx: worker process root 1309 1250 0 01:26 pts/0 00:00:00 grep --color=auto nginx [root@localhost ~]# vi /apps/nginx/conf/nginx.conf user nginx nginx; worker_processes auto; [root@localhost ~]# nginx -t nginx: the configuration file /apps/nginx/conf/nginx.conf syntax is ok nginx: configuration file /apps/nginx/conf/nginx.conf test is successful [root@localhost ~]# nginx -s reload [root@localhost ~]# ps -ef|grep nginx root 1095 1 0 23:38 ? 00:00:00 nginx: master process nginx nginx 1103 1095 0 23:40 ? 00:00:00 nginx: worker process nginx 1104 1095 0 23:40 ? 00:00:00 nginx: worker process root 1107 1073 0 23:41 pts/0 00:00:00 grep --color=auto nginx
The binding does not mean that the current process nginx exclusive one CPU core, but can guarantee that once the process is not run on other core, which greatly reduced
Less nginx worker processes to jump back and forth on a different CPU core, reducing the CPU to the process of resource allocation and recovery as well as memory management, so you can
Effectively improve the nginx server performance.
[the root @ localhost ~] # VI /apps/nginx/conf/nginx.conf User Nginx Nginx; worker_processes Auto; worker_cpu_affinity 00000001 00000010; [the root @ localhost ~] # -t Nginx Nginx: The Configuration File / Apps / Nginx / the conf IS the ok syntax /nginx.conf nginx: the Configuration File /apps/nginx/conf/nginx.conf the Test IS successful [root @ localhost ~] # nginx -s reload [root @ localhost ~] # PS AXO pid, cmd, PSR | Nginx grep 1095 Nginx: Nginx process Master. 1 1113 Nginx: process worker 0 1114 Nginx: process worker. 1 1116 grep --color Nginx 0 = Auto # current can be seen from the above figure process worker working on a different CPU in
Cheng run on which CPU
[root@localhost ~]# yum -y install httpd-tools #下载ab命令 [root@localhost ~]# ps axo pid,cmd,psr |grep nginx 1095 nginx: master process nginx 1 1124 nginx: worker process 0 1125 nginx: worker process 1 1165 grep --color=auto nginx 1 [root@localhost ~]# ab -n1000 -c1000 http://192.168.1.170/index.html This is ApacheBench, Version 2.3 <$Revision: 1430300 $> Copyright 1996 Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/ Licensed to The Apache Software Foundation, http://www.apache.org/ Benchmarking 192.168.1.170 (be patient) Completed 100 requests Completed 200 requests Completed 300 requests Completed 400 requests Completed 500 requests Completed 600 requests Completed 700 requests Completed 800 requests Completed 900 requests Completed 1000 requests Finished 1000 requests Server Software: nginx/1.14.2 Server Hostname: 192.168.1.170 Server Port: 80 Document Path: /index.html Document Length: 612 bytes Concurrency Level: 1000 Time taken for tests: 0.065 seconds Complete requests: 1000 Failed requests: 0 Write errors: 0 Total transferred: 845000 bytes HTML transferred: 612000 bytes Requests per second: 15424.24 [#/sec] (mean) Time per request: 64.833 [ms] (mean) Time per request: 0.065 [ms] (mean, across all concurrent requests) Transfer rate: 12728.01 [Kbytes/sec] received Connection Times (ms) min mean[+/-sd] median max Connect: 0 17 5.8 16 27 Processing: 7 15 6.9 11 30 Waiting: 0 14 7.5 11 30 Total: 17 32 5.7 34 38 Percentage of the requests served within a certain time (ms) 50% 34 66% 35 75% 36 80% 36 90% 36 95% 37 98% 38 99% 38 100% 38 (longest request) [root@localhost ~]# ps axo pid,cmd,psr |grep nginx 1095 nginx: master process nginx 1 1124 nginx: worker process 0 1125 nginx: worker process 0 1168 grep --color=auto nginx 1
The maximum number of concurrent to worker_connections * worker_processes; as a reverse proxy server or load-balancing server time,
The maximum number of concurrent nginx is (worker_connections * worker_processes) / 2; change the default parameter is 1024; 10 may also be provided
Million connections.
# View the default configuration [root @ localhost ~] # CAT /apps/nginx/conf/nginx.conf|grep worker worker_processes Auto; worker_cpu_affinity 00000001 00000010; worker_connections 1024; # default configuration size # settings on each worker can process 100,000 concurrent connections [the root @ localhost ~] -1 worker_connections /apps/nginx/conf/nginx.conf|grep CAT # Events { worker_connections 100000; }