Abstract: A cluster network that includes multiple server nodes is disclosed. Each of the multiple nodes of the cluster network is coupled to first and second expander devices. The cluster network includes a number of storage enclosures, each of which includes dual interface and a number of dual port storage devices. Each of the dual port storage devices are coupled to each interface of the storage enclosure. Each interface is coupled to one of the expander devices. If one of the expander devices fails, the server nodes of the cluster network can access each other peer server node and each storage drive of the cluster network.

Abstract: A system and method for recovering from a drive failure in a storage array is disclosed in which a spare drive array is provided in the event of a failure in one of the active drives of the storage array. The spare drive array is comprised of multiple drives that are comprised in a RAID 0 configuration to reduce the time required to rebuild the content of the failed drive on the spare drive.

Abstract: A system and method is disclosed for identifying the boot targets for an initiator port in point-to-point network, such as a Serial Attached SCSI network. In operation, the initiator port issues a command to an expander that is directly coupled to the initiator port. In response the expander identifies the expander of the network that includes a boot table that identifies the boot addresses associated with each of the initiator port. The boot addresses associated with the initiator port are returned to the initiator port in a ranked order, and the initiator port attempts to boot from the boot address locations until the initiator port successfully boots.

Abstract: A cluster network is disclosed in which of the storage drives of the network are distributed across the nodes of the network. Data is stored to the drives of the network according to a RAID storage methodology and without the necessity for shared external storage.

Abstract: A system and method is disclosed for establishing a mirror configuration in a storage network. The storage network includes two storage drives that are coupled to one another according to a Serial Attached SCSI storage network interface. The storage drives determine whether the storage drives are coupled to one another. If it is determined that the storage drives are coupled to one another, the storage drives are configured in a mirror configuration in which write commands received at one of the drives are mirrored to the other drive.

Abstract: A system and method is disclosed for managing configuration discovery following a loss or acquisition of link synchronization on a link in a network having a SAS architecture with one or more expanders. When a loss or acquisition of link synchronization on a link is identified, the IDENTIFY address frame is examined to determine if a discovery sequence should be performed by the initiator ports of the network. Depending on whether a target port or an initiator port is involved in the loss or acquisition of link synchronization, a broadcast primitive may not be issued, or may only be issued to the initiator ports or expander ports.

Abstract: The size of the spiral gaskets which are formed by winding a sealing strip and a compressible filler strip around a mandrel is controlled by measuring the angle through which the mandrel turns, e.g., by means of an encoder. The lengths of arcs on the periphery of the gasket as it is wound around the mandrel are also measured, e.g., by means of a wheel running on the periphery of the gasket and an encoder counting turns of the wheel. The arcs are defined by the angles through which the mandrel turns. The supply of strip is terminated, e.g., by cutting the strips, as result of an arc achieving a particular length.