Abstract: Techniques for monitoring and sending network connectivity information for connection pathways to host machines is provided. A storage node comprises a plurality of attached network interfaces and is operatively coupled to storage devices that are accessible to the host machines through network interfaces attached to the storage node. Each of the network interfaces send messages to the host machines. The storage node, receives host messages from at least one host machine. The storage node determines connectivity information about connectivity between network interfaces and host machines. The storage node receives a request for data that is stored on the one or more storage devices, from a host machine, at a target network interface, where the target node was chosen by the host machine based upon the connectivity information.

Abstract: Techniques for monitoring and sending network connectivity information for connection pathways to host machines is provided. A storage node comprises a plurality of attached network interfaces and is operatively coupled to storage devices that are accessible to the host machines through network interfaces attached to the storage node. Each of the network interfaces send messages to the host machines. The storage node, receives host messages from at least one host machine. The storage node determines connectivity information about connectivity between network interfaces and host machines. The storage node receives a request for data that is stored on the one or more storage devices, from a host machine, at a target network interface, where the target node was chosen by the host machine based upon the connectivity information.

Abstract: A distributed storage system that dispatches an input/output request is described. In an exemplary embodiment, a storage controller client receives the input/output request, wherein the distributed storage system includes the storage controller client, a plurality of storage controller servers, a plurality of virtual nodes distributed among a plurality of physical nodes, and each of the plurality of physical nodes is hosted on one of the plurality of storage controller servers. The storage controller client further computes a target virtual node for the input/output request, where the target virtual node is one of the plurality of virtual nodes. Using the computed target virtual node, the storage controller client determines a target physical node that corresponds to the target virtual node, where the target physical node is one of the plurality of physical nodes.

Abstract: A distributed storage system that dispatches an input/output request is described. In an exemplary embodiment, a storage controller client receives the input/output request, wherein the distributed storage system includes the storage controller client, a plurality of storage controller servers, a plurality of virtual nodes distributed among a plurality of physical nodes, and each of the plurality of physical nodes is hosted on one of the plurality of storage controller servers. The storage controller client further computes a target virtual node for the input/output request, where the target virtual node is one of the plurality of virtual nodes. Using the computed target virtual node, the storage controller client determines a target physical node that corresponds to the target virtual node, where the target physical node is one of the plurality of physical nodes.

Abstract: A system for recovery of process relationships following node failure within a computer cluster is provided. For relationship recovery, each node maintains set of care relationships. Each relationship is of the form carer cares about care target. Care relationships describe process relations such as parent-child or group leader-group member. Care relationships are stored at the origin node of their care targets. Following node failure, a surrogate origin node is selected. The surviving nodes then cooperate to rebuild vproc structures and care relationships for the processes that originated at the failed node at the surrogate origin node. The surviving nodes then determine which of their own care targets were terminated by the node failure. For each terminated care targets, notifications are sent to the appropriate carers. This allows surviving processes to correctly recover from severed process relationships.