Showing papers on "Shared resource published in 1974"

Interprocessor communication apparatus for a data processing system

Abstract: In a data processing system having independently operating asynchronous processors, apparatus is disclosed which provides for interprocessor synchronization and/or information exchange. Synchronization interlocks and controls are provided for both identifying shared resources of a control processor and an input/output controller (IOC) processor and for obtaining control over these shared resources. If a conflict situation for any one of the shared resources arises, apparatus is disclosed whereby the IOC processor is provided the capability of assuming control over the shared resource even though the central processor has control over it. One of the shared resources is an interprocessor communication register which allows communication of control information between both the central processor and the IOC processor and from the central processor to the peripheral processor over a shared bus.

The PROPHET system and resource sharing.

Abstract: The PROPHET System is a specialized computer resource sponsored by the National Institutes of Health (NTH). The short-term objective for the System is that it be a unique and valuable research tool shared by a broad array of geographically distributed scientists studying chemical/ biological interactions. The long-term goal for the System is that it promote the emergence of a predictive science regarding drugs and drug-related phenomena. This paper discusses this shared research resource and the management considerations associated with its operation and continuing development.

Deadlock prevention in message switched networks

Abstract: The importance of computer networking to achieve resource sharing and data sharing is widely recognized. Local networking to achieve high performance and reliability is inevitable for the future. As cost of logic and memory decreases, the cost of communications resources become increasingly significant and these resources must be increasingly shared. This heightens the possibility of deadlocks due to this sharing. Deadlocks in the ARPA network are examples of this type of deadlock. Deadlock problems due to sharing of the resources allocated by operating systems have been extensively studied. Deadlock problems in computer networks require significantly different approaches due to the distributed control of the network. This paper studies the deadlock problem in this new context; as an example, a simple solution to the ARPA network store-and-forward lockup problem is given.

A model highlighting the security of operating systems

Abstract: The major thrust toward providing secure computing facilities has gone into the design of, or models for, new operating systems. Work directed toward securing current systems has, for the most part, taken the form of penetration attempts. Penetration efforts have led several authors to identify generic weaknesses, but grouping by weakness has not led to formal methods. An approach showing greater promise in identifying trouble spots, as well as characterizing existing operating systems in a more general sense, lies in forming graph models in which nodes are program modules or data structures, and arcs are access or shared resource synchronization paths. A given system should be capable of reduction to a graph of this sort by appropriate analysis of its load modules.

Economies and economics of network computers.

Abstract: Network computers have evolved for reasons of economics and efficiency. A network computer is a set of computer systems which have been interconnected to achieve some common purpose: be it increased reliability, load sharing, or resource sharing. The major goal of most network computers presently in existence or planned is resource sharing (4, 10, 13). The operational and managerial advantages that may be derived from resource sharing and the attendant problems of network resource management are discussed in (14).