Shared disk architecture

Abstract: A method and system for managing data (i.e., objects) that are shared by multiple instances of a shared application program. A shared application program is an application program that is executing simultaneously on multiple computers and that has a copy of data that is being shared by each instance of the application program. In particular, each computer maintains a copy of the shared data. When an instance of the application program modifies the shared data, the modifications are sent to the other computers. Each of these other computers stores the data in its copy of the shared data, and each instance of the application program updates its user interface to reflect the modifications to the shared data. Thus, the users of the shared application program can cooperatively modify and view the shared data. For example, multiple users can be executing a word processing program on their computer and sharing a common document. As one user changes the document, the word processing program updates its copy of the shared data. The changes are then transmitted to the other computers so that their copy of the shared data can be updated. In particular, an object management (OM) system is provided that enables shared application programs to manage their copy of the shared data. Each computer has a copy of the OM system. The OM system, under the direction of the shared application program, manages the adding, deleting, and modifying of the shared data. The OM system also controls the transmitting of modifications to the copy of the shared data to the other computers.

Abstract: A cluster architecture for a highly parallel multiprocessor computer processing system (40) is comprised of one or more clusters of tightly-coupled, high-speed processors (10) capable of both vector and scalar parallel processing that can symmetrically access shared resources (12) associated with the cluster, as well as the shared resources associated with other clusters.

Abstract: A new, large scale multiprocessor architecture is presented in this paper. The architecture consists of hierarchies of shared buses and caches. Extended versions of shared bus multicache coherency protocols are used to maintain coherency among all caches in the system. After explaining the basic operation of the strict hierarchical approach, a clustered system is introduced which distributes the memory among groups of processors. Results of simulations are presented which demonstrate that the additional coherency protocol overhead introduced by the clustered approach is small. The simulations also show that a 128 processor multiprocessor can be constructed using this architecture which will achieve a substantial fraction of its peak performance. Finally, an analytic model is used to explore systems too large to simulate (with available hardware). The model indicates that a system of over 1000 usable MIPS can be constructed using high performance microprocessors.

Abstract: A system and method for recovering from failures in the disk access path of a clustered computing system Each node of the clustered computing system is provided with proxy software for handling physical disk access requests from applications executing on the node and for directing the disk access requests to an appropriate server to which the disk is physically attached The proxy software on each node maintains state information for all pending requests originating from that node In response to detection of a failure along the disk access path, the proxy software on all of the nodes directs all further requests for disk access to a secondary node physically attached to the same disk