out of socket memory調整
主要是調整TCP緩衝區
Increasing the TCP send and receive buffers will increase the performance a lot if (and only if) you have a lot of large files to send.
增加TCP 發送和接收緩衝區對效能有很大的提高,在你有大檔案發送的情況下
net.ipv4.tcp_wmem = 4096 65536 524288
net.core.wmem_max = 1048576
If you have a lot of large file uploads, increasing the receive buffers will help.
如果你有大檔案上傳,增大接收緩衝區對給你帶來協助
net.ipv4.tcp_rmem = 4096 87380 524288
net.core.rmem_max = 1048576
# These ensure that TIME_WAIT ports either get reused or closed fast.
以下參數確保TIME_WAIT 快速關閉
net.ipv4.tcp_fin_timeout = 1 ##開啟fin_wait 快速關閉參數
net.ipv4.tcp_tw_recycle = 1 ##快速回收fin_wait參數
設定核心TCP的相關參數
# TCP memory
net.core.rmem_max = 16777216 ##核心最大接收緩衝 16M
net.core.rmem_default = 16777216 ##核心黙認接收緩衝 16M
net.core.netdev_max_backlog = 262144 ##2M
net.core.somaxconn = 262144 ##2M
net.ipv4.tcp_syncookies = 1 ##開啟SYN_COOKIE
net.ipv4.tcp_max_orphans = 262144 ##2M
net.ipv4.tcp_max_syn_backlog = 262144 ##SYN 隊列大小 2M
net.ipv4.tcp_synack_retries = 2 ##SYN_ACK重試時間 2秒
net.ipv4.tcp_syn_retries = 2 ## SYN 重試時間 2秒
# you shouldn't be using conntrack on a heavily loaded server anyway, but these are
# suitably high for our uses, insuring that if conntrack gets turned on, the box doesn’t die
如果用到IPTABLES 同樣在增長ip_conntrack 的大小,這個是iptables 所管理的狀態數目
net.ipv4.ip_conntrack_max = 1048576
net.ipv4.netfilter.ip_conntrack_max = 1048576
原文:
http://hi.baidu.com/lewutian/item/b7fe3cc56f966345a8ba94c2