Abstract: Various embodiments of a cluster of application servers in a networked computer system in which each application server includes a separate transaction manager instance are disclosed. In one embodiment, a separate transaction manager instance may be used to coordinate transactions initiated by an application component executing in process with the application server for a node of a cluster. If each transaction manager instance manages only transactions initiated within its own application server, then the loss of a single transaction manager instance may have no detrimental effect on transactions initiated in other nodes of the cluster. Further, if each transaction manager instance maintains its transaction logs in highly available memory accessible to all nodes of the cluster, in flight transactions of a failed node may be recovered by an instance in a different node.

Abstract: A system and method for a highly available timer service may include a timer database operable to store information on one or more timers. The system may further include a plurality of servers in a cluster, each server containing a failure detection service, a timer service and one or more applications. In one embodiment, each application may be operable to request and process one or more timer notifications from the timer service contained in the same server. Each failure detection service may also be operable to detect a failure of other servers in the cluster. In addition, each timer service may be operable to retrieve information on the one or more timers from the timer database, and to provide a timer notification to any application in the cluster in response to a notification from the failure notification service that a server has failed.

Abstract: A system and method for selectively checkpointing application components. In one embodiment this method may include deploying a plurality of application components on a server, determining checkpoint information for each application component on the server, and selectively checkpointing each component during execution of each component according to the checkpoint information for each component. In one embodiment the checkpoint information may be provided by a checkpoint selection heuristic, which makes checkpoint selections based on a description of one or more methods of the application component. In one embodiment the checkpoint information may be provided by a deployment descriptor, which may have multiple overriding tiers to facilitate precise selection of methods, components and modules for checkpointing. In one embodiment the checkpoint and activation process may include callback routines which prepare an application component for checkpointing and activation.