Difference between revisions of "TCPHA"

From LVSKB
Jump to: navigation, search
 
m
 
(5 intermediate revisions by the same user not shown)
Line 1: Line 1:
TCPHA (TCP HAndoff) implements an architecture for scalable content-aware request distribution in cluster-base servers. It implements Kernel layer-7 switching based on TCP Handoff protocol for the Linux operating system. Since the overhead of layer-7 switching in user-space is very high, it is good to implement it inside the kernel in order to avoid the overhead of context switching and memory copying between user-space and kernel-space, furthermore, the responses are sent directly to clients, not passing the dispatcher, which will greatly improve the performance of cluster.
+
TCPHA ('''TCP''' '''HA'''ndoff) implements an architecture for scalable content-aware request distribution in cluster-base servers. It implements Kernel layer-7 switching based on [[TCP Handoff]] protocol for the Linux operating system. Since the overhead of layer-7 switching in user-space is very high, it is good to implement it inside the kernel in order to avoid the overhead of [[context switch]]ing and memory copying between user-space and kernel-space. Furthermore, the responses are sent directly back to clients without passing through the dispatcher, which can greatly improve the performance of cluster.
  
See http://dragon.linux-vs.org/~dragonfly/
+
See http://dragon.linux-vs.org/~dragonfly/htm/tcpha.htm
 +
 
 +
[[Category:LVS Software]]
 +
[[Category:Terms and Abbreviations]]

Latest revision as of 12:52, 15 November 2006

TCPHA (TCP HAndoff) implements an architecture for scalable content-aware request distribution in cluster-base servers. It implements Kernel layer-7 switching based on TCP Handoff protocol for the Linux operating system. Since the overhead of layer-7 switching in user-space is very high, it is good to implement it inside the kernel in order to avoid the overhead of context switching and memory copying between user-space and kernel-space. Furthermore, the responses are sent directly back to clients without passing through the dispatcher, which can greatly improve the performance of cluster.

See http://dragon.linux-vs.org/~dragonfly/htm/tcpha.htm