
ab -kc 250 -n 1000 -H "Accept-EnCoding: gzip,deflate" https://www.mysite.com/Server Software: Nginx/1.0.15Server Hostname: www.mysite.comServer Port: 443SSL/TLS Protocol: TLSv1/SSLv3,DHE-RSA-AES256-SHA,2048,256document Path: /document Length: 4258 bytesConcurrency Level: 250Time taken for tests: 16.498 secondsComplete requests: 1000HTML transferred: 4258000 bytesRequests per second: 60.61 [#/sec] (mean)Time per request: 4124.432 [ms] (mean)Time per request: 16.498 [ms] (mean,across all concurrent requests)Transfer rate: 282.83 [Kbytes/sec] receivedConnection Times (ms) min mean[+/-sd] median maxConnect: 302 1801 866.3 1623 3583Processing: 183 1993 867.9 1873 7050Waiting: 183 1880 864.2 1743 7036Total: 1397 3794 1389.0 3558 10580Percentage of the requests served within a certain time (ms) 50% 3558 66% 4012 75% 4179 80% 4712 90% 4947 95% 6411 98% 8376 99% 8380 100% 10580 (longest request)
呸.也许我错了,但我觉得我应该能够获得远高于每秒60个请求的数量,并且每次请求的时间要低得多于4.1秒.
该应用程序位于c1.xlarge EC2实例(7 GB内存20 EC2计算单元,8个虚拟核心,每个2.5 EC2计算单元,1690 GB实例存储64位平台,I / O性能:高). ApacheBench基准测试运行的站点根目录正在进行 *** 作缓存,甚至不会触及数据库.对它的单个请求如下所示:
Started GET "/" for ##.###.###.## at Thu Apr 19 13:05:50 +0000 2012 Processing by OneOfMyControllers#index as HTML Read fragment vIEws/www.mysite.com/index (1.1ms) Completed 200 OK in 2ms
该机器正在运行Ubuntu 11.04,我正在使用Rails 3.1,Ruby Enterprise Edition,Passenger和Nginx.我的Nginx配置看起来像这样:
worker_processes 8;worker_rlimit_nofile 65535;events { worker_connections 4096;}http { passenger_root /usr/lib/ruby/gems/1.8/gems/passenger-3.0.8; passenger_ruby /usr/bin/ruby1.8; rails_spawn_method smart; rails_app_spawner_IDle_time 0; rails_framework_spawner_IDle_time 0; passenger_max_pool_size 60; passenger_pool_IDle_time 1000; ssl_session_cache shared:SSL:3m; # 3MB can hold about 12k SSL cache sessions ssl_session_timeout 5m; # average user spends 5 minutes on our site include mime.types; default_type application/octet-stream; sendfile on; tcp_nopush on; # useful with the sendfile option tcp_nodelay off; keepalive_timeout 10; send_timeout 10; clIEnt_body_timeout 10; clIEnt_header_timeout 10; gzip on; gzip_static on; gzip_disable "MSIE [1-6]\."; gzip_comp_level 4; gzip_buffers 16 4k; gzip_min_length 1000; gzip_proxIEd any; gzip_vary on; gzip_types text/plain text/CSS application/Json application/x-JavaScript text/xml application/xml application/xml+RSS text/JavaScript; gzip_http_version 1.0; # Amazon CloudFront uses http/1.0 include /opt/Nginx/conf/sites-enabled/*;} 在/opt/Nginx/conf/sites-enabled/mysite.com里面,我得到了:
server { Listen 80; server_name mysite.com; rewrite ^/(.*) http://www.mysite.com/ permanent;}server { Listen 443; ssl on; server_name mysite.com; ssl_certificate /opt/mysite/ssl/wildcard_gandi_2012/combined-certification.crt; ssl_certificate_key /opt/mysite/ssl/wildcard_gandi_2012/monserveur.key; ssl_protocols SSLv3 TLSv1; ssl_ciphers HIGH:!ADH:!MD5; ssl_prefer_server_ciphers on; rewrite ^/(.*) https://www.mysite.com/ permanent;}server { Listen 80; server_name *.mysite.com www.mysite.com; root /opt/mysite/public; passenger_enabled on; access_log /opt/mysite/log/Nginx_access.log; error_log /opt/mysite/log/Nginx_error.log; # Set the maximum file upload size to 25 MB. clIEnt_max_body_size 25M;}server { Listen 443; ssl on; server_name *.mysite.com www.mysite.com; ssl_certificate /opt/mysite/ssl/wildcard_gandi_2012/combined-certification.crt; ssl_certificate_key /opt/mysite/ssl/wildcard_gandi_2012/monserveur.key; ssl_protocols SSLv3 TLSv1; ssl_ciphers HIGH:!ADH:!MD5; ssl_prefer_server_ciphers on; root /opt/mysite/public; passenger_enabled on; access_log /opt/mysite/log/Nginx_access.log; error_log /opt/mysite/log/Nginx_error.log; # Set the maximum file upload size to 25 MB. clIEnt_max_body_size 25M; # Asset caching location ~ ^/(assets)/ { root /opt/mysite/public; gzip_static on; access_log off; expires 1y; add_header Cache-Control public; add_header Last-ModifIEd ""; add_header ETag ""; break; }} 知道为什么我不能获得超过60个请求/秒?我有什么明显的东西可以忽略吗?
解决方法 我怀疑你的一个重大问题是https.我刚刚运行了一些比较基准测试,对我来说,他们在这样一个非常基本的页面上显示速度降低了2-4倍.如果您点击http页面,您的指标会增加吗?
还要检查您的乘客状态(rvmsudo passenger-status)以检查是否已加载所有进程并均匀地在其中分配请求.
总结以上是内存溢出为你收集整理的ruby-on-rails – 大型机器上的Rails应用程序只能获得60个请求/秒的基准测试结果全部内容,希望文章能够帮你解决ruby-on-rails – 大型机器上的Rails应用程序只能获得60个请求/秒的基准测试结果所遇到的程序开发问题。
如果觉得内存溢出网站内容还不错,欢迎将内存溢出网站推荐给程序员好友。
欢迎分享,转载请注明来源:内存溢出
微信扫一扫
支付宝扫一扫
评论列表(0条)