2 # IP Virtual Server configuration
5 tristate "IP virtual server support"
6 depends on NET && INET && NETFILTER
8 IP Virtual Server support will let you build a high-performance
9 virtual server based on cluster of two or more real servers. This
10 option must be enabled for at least one of the clustered computers
11 that will take care of intercepting incoming connections to a
12 single IP address and scheduling them to real servers.
14 Three request dispatching techniques are implemented, they are
15 virtual server via NAT, virtual server via tunneling and virtual
16 server via direct routing. The several scheduling algorithms can
17 be used to choose which server the connection is directed to,
18 thus load balancing can be achieved among the servers. For more
19 information and its administration program, please visit the
20 following URL: <http://www.linuxvirtualserver.org/>.
22 If you want to compile it in kernel, say Y. To compile it as a
23 module, choose M here. If unsure, say N.
28 bool "IPv6 support for IPVS (DANGEROUS)"
29 depends on EXPERIMENTAL && (IPV6 = y || IP_VS = IPV6)
31 Add IPv6 support to IPVS. This is incomplete and might be dangerous.
36 bool "IP virtual server debugging"
38 Say Y here if you want to get additional messages useful in
39 debugging the IP virtual server code. You can change the debug
40 level in /proc/sys/net/ipv4/vs/debug_level
43 int "IPVS connection table size (the Nth power of 2)"
47 The IPVS connection hash table uses the chaining scheme to handle
48 hash collisions. Using a big IPVS connection hash table will greatly
49 reduce conflicts when there are hundreds of thousands of connections
52 Note the table size must be power of 2. The table size will be the
53 value of 2 to the your input number power. The number to choose is
54 from 8 to 20, the default number is 12, which means the table size
55 is 4096. Don't input the number too small, otherwise you will lose
56 performance on it. You can adapt the table size yourself, according
57 to your virtual server application. It is good to set the table size
58 not far less than the number of connections per second multiplying
59 average lasting time of connection in the table. For example, your
60 virtual server gets 200 connections per second, the connection lasts
61 for 200 seconds in average in the connection table, the table size
62 should be not far less than 200x200, it is good to set the table
65 Another note that each connection occupies 128 bytes effectively and
66 each hash entry uses 8 bytes, so you can estimate how much memory is
69 comment "IPVS transport protocol load balancing support"
71 config IP_VS_PROTO_TCP
72 bool "TCP load balancing support"
74 This option enables support for load balancing TCP transport
75 protocol. Say Y if unsure.
77 config IP_VS_PROTO_UDP
78 bool "UDP load balancing support"
80 This option enables support for load balancing UDP transport
81 protocol. Say Y if unsure.
83 config IP_VS_PROTO_AH_ESP
87 config IP_VS_PROTO_ESP
88 bool "ESP load balancing support"
89 select IP_VS_PROTO_AH_ESP
91 This option enables support for load balancing ESP (Encapsulation
92 Security Payload) transport protocol. Say Y if unsure.
95 bool "AH load balancing support"
96 select IP_VS_PROTO_AH_ESP
98 This option enables support for load balancing AH (Authentication
99 Header) transport protocol. Say Y if unsure.
101 comment "IPVS scheduler"
104 tristate "round-robin scheduling"
106 The robin-robin scheduling algorithm simply directs network
107 connections to different real servers in a round-robin manner.
109 If you want to compile it in kernel, say Y. To compile it as a
110 module, choose M here. If unsure, say N.
113 tristate "weighted round-robin scheduling"
115 The weighted robin-robin scheduling algorithm directs network
116 connections to different real servers based on server weights
117 in a round-robin manner. Servers with higher weights receive
118 new connections first than those with less weights, and servers
119 with higher weights get more connections than those with less
120 weights and servers with equal weights get equal connections.
122 If you want to compile it in kernel, say Y. To compile it as a
123 module, choose M here. If unsure, say N.
126 tristate "least-connection scheduling"
128 The least-connection scheduling algorithm directs network
129 connections to the server with the least number of active
132 If you want to compile it in kernel, say Y. To compile it as a
133 module, choose M here. If unsure, say N.
136 tristate "weighted least-connection scheduling"
138 The weighted least-connection scheduling algorithm directs network
139 connections to the server with the least active connections
140 normalized by the server weight.
142 If you want to compile it in kernel, say Y. To compile it as a
143 module, choose M here. If unsure, say N.
146 tristate "locality-based least-connection scheduling"
148 The locality-based least-connection scheduling algorithm is for
149 destination IP load balancing. It is usually used in cache cluster.
150 This algorithm usually directs packet destined for an IP address to
151 its server if the server is alive and under load. If the server is
152 overloaded (its active connection numbers is larger than its weight)
153 and there is a server in its half load, then allocate the weighted
154 least-connection server to this IP address.
156 If you want to compile it in kernel, say Y. To compile it as a
157 module, choose M here. If unsure, say N.
160 tristate "locality-based least-connection with replication scheduling"
162 The locality-based least-connection with replication scheduling
163 algorithm is also for destination IP load balancing. It is
164 usually used in cache cluster. It differs from the LBLC scheduling
165 as follows: the load balancer maintains mappings from a target
166 to a set of server nodes that can serve the target. Requests for
167 a target are assigned to the least-connection node in the target's
168 server set. If all the node in the server set are over loaded,
169 it picks up a least-connection node in the cluster and adds it
170 in the sever set for the target. If the server set has not been
171 modified for the specified time, the most loaded node is removed
172 from the server set, in order to avoid high degree of replication.
174 If you want to compile it in kernel, say Y. To compile it as a
175 module, choose M here. If unsure, say N.
178 tristate "destination hashing scheduling"
180 The destination hashing scheduling algorithm assigns network
181 connections to the servers through looking up a statically assigned
182 hash table by their destination IP addresses.
184 If you want to compile it in kernel, say Y. To compile it as a
185 module, choose M here. If unsure, say N.
188 tristate "source hashing scheduling"
190 The source hashing scheduling algorithm assigns network
191 connections to the servers through looking up a statically assigned
192 hash table by their source IP addresses.
194 If you want to compile it in kernel, say Y. To compile it as a
195 module, choose M here. If unsure, say N.
198 tristate "shortest expected delay scheduling"
200 The shortest expected delay scheduling algorithm assigns network
201 connections to the server with the shortest expected delay. The
202 expected delay that the job will experience is (Ci + 1) / Ui if
203 sent to the ith server, in which Ci is the number of connections
204 on the ith server and Ui is the fixed service rate (weight)
207 If you want to compile it in kernel, say Y. To compile it as a
208 module, choose M here. If unsure, say N.
211 tristate "never queue scheduling"
213 The never queue scheduling algorithm adopts a two-speed model.
214 When there is an idle server available, the job will be sent to
215 the idle server, instead of waiting for a fast one. When there
216 is no idle server available, the job will be sent to the server
217 that minimize its expected delay (The Shortest Expected Delay
218 scheduling algorithm).
220 If you want to compile it in kernel, say Y. To compile it as a
221 module, choose M here. If unsure, say N.
223 comment 'IPVS application helper'
226 tristate "FTP protocol helper"
227 depends on IP_VS_PROTO_TCP
229 FTP is a protocol that transfers IP address and/or port number in
230 the payload. In the virtual server via Network Address Translation,
231 the IP address and port number of real servers cannot be sent to
232 clients in ftp connections directly, so FTP protocol helper is
233 required for tracking the connection and mangling it back to that of
236 If you want to compile it in kernel, say Y. To compile it as a
237 module, choose M here. If unsure, say N.