This paper describes a method to recast the elements of resolution and unification in terms of reduction rules. The term ' epsilon -reduction' has been coined to emphasize similarities to beta -reduction of th...
详细信息
ISBN:
(纸本)0444879870
This paper describes a method to recast the elements of resolution and unification in terms of reduction rules. The term ' epsilon -reduction' has been coined to emphasize similarities to beta -reduction of the lambda -calculus. epsilon -reduction plays a corresponding role in the predicate calculus. It eliminates an existentially quantified variable by substituting for it an expression to which it has been found to be equal. A combination of function and logic programming is achieved as an extension of the reduction rules implemented. The extension comprises two reduction rules. Their implementation and semantics, and the mechanisms to obtain correct scoping of variables, is described. An example in terms of rewrites is presented.
Several large production programs were modified to efficiently use a parallel computer in a time-shared production environment. These programs included the circuit simulator SPICE 2G5 and the finite-element structure ...
详细信息
ISBN:
(纸本)0818606924
Several large production programs were modified to efficiently use a parallel computer in a time-shared production environment. These programs included the circuit simulator SPICE 2G5 and the finite-element structure deformation program DYNA3D. The resulting parallel performance is reported along with a discussion of the parallelization methodologies and implementation issues in performing the modifications. The implications of parallel processes sharing resources with other users of the computer system are also considered.
The author describes the design of a generic sorting package, showing that Ada reuse can be accomplished during, and even prior to, coding. Some key technical issues are identified in reuse. These issues are of genera...
详细信息
ISBN:
(纸本)0818607181
The author describes the design of a generic sorting package, showing that Ada reuse can be accomplished during, and even prior to, coding. Some key technical issues are identified in reuse. These issues are of general interest to a software engineer, as they focus on program design and implementation. The conclusions are based on a generic program unit which includes array sorting algorithms adapted and generalized to work as a generic program unit. Any array component type except limited types may be directly sorted;limited types may be indirectly sorted. Examples of direct and indirect sorting are presented. This package may also be used in conjunction with a merging package to sort data residing on external memory devices, in effect allowing files of arbitrary length to be sorted by means of an extended operation.
We describe in this short paper a total systems design of an interfacing system for a PC. The hardware interface (the Universal Interface) uses a standard I/O rack which can be interfaced to most types of PC's via...
详细信息
We describe in this short paper a total systems design of an interfacing system for a PC. The hardware interface (the Universal Interface) uses a standard I/O rack which can be interfaced to most types of PC's via a personality module. A special suite of subroutines called BASIL provides the link between the user's BASIC program and the Universal Interface.
The merits or otherwise of using the International Telephone and Telegraph Consultative Committee (CCITT) Specification and Description Language (SDL) for specification purposes has received much discussion in the lit...
详细信息
ISBN:
(纸本)0858253062
The merits or otherwise of using the International Telephone and Telegraph Consultative Committee (CCITT) Specification and Description Language (SDL) for specification purposes has received much discussion in the literature. There has, however, been little discussion of the methods of developing program code from such specifications nor has there been discussion of the advantages or disadvantages of producing code in this way. This paper describes experience with implementing software from an SDL specification and discusses various design approaches and the problems encountered.
This paper seeks to demonstrate that experience of designing sequential programs using data abstraction may usefully be transferred to the design of concurrent programs. Solutions to an example derived both by functio...
详细信息
ISBN:
(纸本)0080334504
This paper seeks to demonstrate that experience of designing sequential programs using data abstraction may usefully be transferred to the design of concurrent programs. Solutions to an example derived both by functional decomposition and by data decomposition are compared with a concurrent solution. A correspondence is exhibited between the sequential concurrent solution. A change to the initial concurrent solution is proposed, once the relationship to data abstraction is recognised, suggesting that by viewing the concurrent solution in this way it is possible to find improved solutions.
The time complexity of sorting n elements using p greater than equivalent to n processors on L. G. Valiant's (1975) parallel comparison tree model is considered. It is shown that this time complexity is THETA (log...
详细信息
ISBN:
(纸本)0818607408
The time complexity of sorting n elements using p greater than equivalent to n processors on L. G. Valiant's (1975) parallel comparison tree model is considered. It is shown that this time complexity is THETA (log n/log(1 plus p/n)). For every fixed time k it is shown that OMEGA (n**1** plus **1 **/ **k ) comparisons are required and that there exists a randomized algorithm for comparison sort in time k with an expected number of O(n**1** plus **1 **k comparisons. This implies that for every fixed k, any deterministic comparison sort algorithm must be asymptotically worse than this randomized algorithm. It is shown that 'approximate sorting' in time 1 requires asymptotically more than n log n processors.
This paper deals with the most important kinds of parallelism that can occur in logic programs, OR parallelism and AND parallelism. To explore parallelism we use a data flow model, where the rules and facts of the log...
详细信息
ISBN:
(纸本)3540164421
This paper deals with the most important kinds of parallelism that can occur in logic programs, OR parallelism and AND parallelism. To explore parallelism we use a data flow model, where the rules and facts of the logic programs are represented as graphs. Beginning with a basic model for OR parallel execution of logic programs, where all the subgoals of a rule are persued in sequential order, we give an extended model, where we try to detect and exploit independencies among subgoals during the execution of a rule (dynamic AND parallelism). The prime when extending the basic model was to improve execution time. The sizes of the graphs remain in a tolerable range.
The Personal Applications Manager (PAM) on the Portable and Portable Plus computers evolved from the original PAM for the HP 150 Touchscreen computer. PAM is designed to provide a novice computer user access to most o...
详细信息
The Personal Applications Manager (PAM) on the Portable and Portable Plus computers evolved from the original PAM for the HP 150 Touchscreen computer. PAM is designed to provide a novice computer user access to most of the features of MS-DOS without forcing the user to learn all of its various commands. PAM provides a file manager, which can perform several file maintenance functions, such as formatting discs, creating directories, and deleting, copying, and renaming files. Using PAM, variables such as the size of system RAM, the current font, and the printer and data communications interfaces can be set and changed according to the needs of the user, in a manner that is transparent to an application program.
It is claimed that no one has been able to exploit the growing library of engineering and scientific programs optimized for the Cray - that is, until now. With the introduction of the SCS-40 from Scientific computer S...
详细信息
It is claimed that no one has been able to exploit the growing library of engineering and scientific programs optimized for the Cray - that is, until now. With the introduction of the SCS-40 from Scientific computersystems, it has become possible to duplicate the X-MP instruction set. In other words, the new machine is the first minisuper-computer that is fully compatible with the Cray.
暂无评论