### Eclipse Workspace Patch 1.0 #P docs-cluster_guide Index: en-US/Clustering_Guide_Concepts.xml =================================================================== --- en-US/Clustering_Guide_Concepts.xml (revision 90941) +++ en-US/Clustering_Guide_Concepts.xml (working copy) @@ -5,11 +5,9 @@ Clustering Concepts - In the next section, we discuss basic concepts behind JBoss's clustering + In the next section, we discuss basic concepts behind JBoss' clustering services. It is helpful that you understand these concepts before reading - the rest of the Clustering Guide. Clustering configurations for specific - types of applications are covered beginning with the next chapter. - + the rest of the Clustering Guide.
Cluster Definition @@ -18,44 +16,53 @@ toward a common goal. In a JBoss Application Server cluster (also known as a “partition”), a node is an JBoss Application Server instance. Communication between the nodes is handled by the JGroups group communication - library, with a JGroups Channel providing the core functionality of tracking + library, with a JGroups Channel providing the core functionality of tracking who is in the cluster and reliably exchanging messages between the cluster members. JGroups channels with the same configuration and name have the ability to dynamically discover each other and form a group. This is why simply executing “run -c all” on two AS instances on the same network is - enough for them to form a cluster – each AS starts a Channel (actually, + enough for them to form a cluster – each AS starts a Channel (actually, several) with the same default configuration, so they dynamically discover each other and form a cluster. Nodes can be dynamically added to or removed - from clusters at any time, simply by starting or stopping a Channel with a + from clusters at any time, simply by starting or stopping a Channel with a configuration and name that matches the other cluster members. - - In summary, a JBoss cluster is a set of AS server instances each of which - is running an identically configured and named JGroups Channel. - On the same AS instance, different services can create their own Channel. - In a default 5.0.x AS, four different services create channels – the web - session replication service, the EJB3 SFSB replication service, the EJB3 - entity caching service, and a core general purpose clustering service known - as HAPartition. In order to differentiate these channels, each must have a - unique name, and its configuration must match its peers yet differ from the - other channels. + On the same AS instance, different services can create their own Channel, + and sometimes more than one. In a standard startup of the AS 5 all configuration, + two different services create a total of four different channels – JBoss + Messaging creates two and a core general purpose clustering service known + as HAPartition creates two more. If you deploy clustered web applications, + clustered EJB3 SFSBs or a clustered JPA/Hibernate entity cache, additional + channels will be created. The channels the AS connects can be divided into + three broad categories: a general purpose channel used by the HAPartition + service, channels created by JBoss Cache for special purpose + caching and cluster wide state replication, and two channels used by + JBoss Messaging. - So, if you go to two AS 5.0.x instances and execute run -c all, - the channels will discover each other and you'll have a conceptual - cluster. It's easy to think of this as a two node - cluster, but it's important to understand that you really have 4 channels, - and hence 4 two node clusters. + So, if you go to two AS 5.x instances and execute run -c all, + the channels created on each server will discover each other and you'll have + a conceptual cluster. It's easy to think of this as a two node + cluster, but it's important to understand that you really have multiple channels, + and hence multiple two node clusters. With JBoss AS, it's the services that + form clusters, not the servers. - On the same network, even for the same service, we may have different clusters. + On the same network, you may have different sets of servers whose + services wish to cluster. shows an example network of JBoss - server instances divided into three clusters, with the third cluster only + server instances divided into three sets, with the third set only having one node. This sort of topology can be set up simply by configuring the AS instances such that within a set of nodes meant to form a cluster the - Channel configurations and names match while they differ from any other - channels on the same network. + channel configurations and names match while they differ from any other + channels on the same network. The AS tries to make this is easy as possible, + such that servers that are meant to cluster only need to have the same values + passed on the command line to the -g (partition name) and + -u (multicast address) startup switches. For each set of + servers, different values should be chosen. The sections on “JGroups Configuration” and + “Isolating JGroups Channels” cover in detail how to configure the AS such + that desired peers find each other and unwanted peers do not.
Clusters and server nodes @@ -64,23 +71,13 @@
- - - The section on “JGroups Configuration” and on “Isolating JGroups Channels” - covers in detail how to configure Channels such that desired peers find - each other and unwanted peers do not. As mentioned above, by default JBoss - AS uses four separate JGroups Channels. These can be divided into two - broad categories: the Channel used by the general purpose HAPartition - service, and three Channels created by JBoss Cache for special purpose - caching and cluster wide state replication. -
Service Architectures - The clustering topography defined by the HAPartition - MBean on each node is of great importance to system administrators. But for - most application developers, you are probably more concerned about the cluster + The clustering topography defined by the JGroups configuration on each + node is of great importance to system administrators. But for + most application developers, the greater concern is probably the cluster architecture from a client application's point of view. Two basic clustering architectures are used with JBoss AS: client-side interceptors (a.k.a smart proxies or stubs) and external load balancers. Which architecture your @@ -126,18 +123,19 @@ - +
External Load Balancer Architecture - Other JBoss services, in particular the HTTP-based services, do not - require the client to download anything. The client (e.g., a web browser) - sends in requests and receives responses directly over the wire using - to certain communication protocols (e.g., the HTTP protocol). In this + The HTTP-based JBoss services do not require the client to download + anything. The client (e.g., a web browser) sends in requests and receives + responses directly over the wire using the HTTP protocol). In this case, an external load balancer is required to process all requests and dispatch them to server nodes in the cluster. The client only needs to know how to contact the load balancer; it has no knowledge of the JBoss AS @@ -171,43 +169,88 @@
Load-Balancing Policies - Both the JBoss client-side interceptor (stub) and load balancer use load balancing policies to determine which server node to which node a new request should be sent. In this section, let's go over the load balancing policies available in JBoss AS. + Both the JBoss client-side interceptor (stub) and load balancer use load balancing policies to determine to which server node a new request should be sent. In this section, let's go over the load balancing policies available in JBoss AS.
Client-side interceptor architecture - In JBoss 5.0.0, the following load balancing options are available when the client-side interceptor architecture is used. The client-side stub maintains a list of all nodes providing the target service; the job of the load balance policy is to pick a node from this list for each request. + In JBoss AS 5, the following load balancing options are available when the client-side interceptor architecture is used. The client-side stub maintains a list of all nodes providing the target service; the job of the load balance policy is to pick a node from this list for each request. + Each policy has two implementation classes, one meant for use by legacy services like EJB2 that use the legacy + detached invoker architecture, and the other meant for services like EJB3 that use AOP-based invocations. - Round-Robin (org.jboss.ha.framework.interfaces.RoundRobin): each call is dispatched to a new node, proceeding sequentially through the list of nodes. The first target node is randomly selected from the list. + Round-Robin: each call is dispatched to a new node, proceeding sequentially through the list of nodes. The first target node is randomly selected from the list. + Implemented by org.jboss.ha.framework.interfaces.RoundRobin (legacy) and org.jboss.ha.client.loadbalance.RoundRobin (EJB3). - Random-Robin (org.jboss.ha.framework.interfaces.RandomRobin): for each call the target node is randomly selected from the list. + Random-Robin: for each call the target node is randomly selected from the list. + Implemented by org.jboss.ha.framework.interfaces.RoundRobin (legacy) and org.jboss.ha.client.loadbalance.RoundRobin (EJB3). - - - - First Available (org.jboss.ha.framework.interfaces.FirstAvailable): one of the available target nodes is elected as the main target and is thereafter used for every call; this elected member is randomly chosen from the list of members in the cluster. When the list of target nodes changes (because a node starts or dies), the policy will choose a new target node unless the currently elected node is still available. Each client-side stub elects its own target node independently of the other stubs, so if a particular client downloads two stubs for the same target service (e.g., an EJB), each stub will independently pick its target. This is an example of a policy that provides “session affinity” or “sticky sessions”, since the target node does not change once established. - - + + + + First Available: one of the available target nodes is elected as the main target and is thereafter used for every call; this elected member is randomly chosen from the list of members in the cluster. When the list of target nodes changes (because a node starts or dies), the policy will choose a new target node unless the currently elected node is still available. Each client-side proxy elects its own target node independently of the other proxies, so if a particular client downloads two proxies for the same target service (e.g., an EJB), each proxy will independently pick its target. This is an example of a policy that provides “session affinity” or “sticky sessions”, since the target node does not change once established. + Implemented by org.jboss.ha.framework.interfaces.FirstAvailable (legacy) and org.jboss.ha.client.loadbalance.aop.FirstAvailable (EJB3). + + - + - First Available Identical All Proxies (org.jboss.ha.framework.interfaces.FirstAvailableIdenticalAllProxies): has the same behaviour as the "First Available" policy but the elected target node is shared by all stubs in the same client-side VM that are associated with the same target service. So if a particular client downloads two stubs for the same target service (e.g. an EJB), each stub will use the same target. + First Available Identical All Proxies: has the same behaviour as the "First Available" policy but the elected target node is shared by all proxies in the same client-side VM that are associated with the same target service. So if a particular client downloads two proxies for the same target service (e.g. an EJB), each proxy will use the same target. + Implemented by org.jboss.ha.framework.interfaces.FirstAvailableIdenticalAllProxies (legacy) and org.jboss.ha.client.loadbalance.aop.FirstAvailableIdenticalAllProxies (EJB3). + - Each of the above is an implementation of the org.jboss.ha.framework.interfaces.LoadBalancePolicy interface; users are free to write their own implementation of this simple interface if they need some special behavior. In later sections we'll see how to configure the load balance policies used by different services. + New in JBoss 5 are a set of "TransactionSticky" load balance policies. + These extend the standard policies above to add behavior such that + all invocations that occur within the scope of a transaction are + routed to the same node (if that node still exists). + These are based on the legacy detached invoker architecture, so they + are not available for AOP-based services like EJB3. + + + + + Transaction-Sticky Round-Robin: Transaction-sticky variant of Round-Robin. + Implemented by org.jboss.ha.framework.interfaces.TransactionStickyRoundRobin. + + + + + + Transaction-Sticky Random-Robin: Transaction-sticky variant of Random-Robin. + Implemented by org.jboss.ha.framework.interfaces.TransactionStickyRandomRobin. + + + + + Transaction-Sticky First Available: Transaction-sticky variant of First Available. + Implemented by org.jboss.ha.framework.interfaces.TransactionStickyFirstAvailable. + + + + + + Transaction-Sticky First Available Identical All Proxies: Transaction-sticky variant of First Available Identical All Proxies. + Implemented by org.jboss.ha.framework.interfaces.TransactionStickyFirstAvailableIdenticalAllProxies. + + + + + + Each of the above is an implementation of a simple interface; users are free to write their own implementations if they need some special behavior. In later sections we'll see how to configure the load balance policies used by different services.
+
External load balancer architecture