Queueing network models of multiprogrammingsystems with memory constraints and multiple classes of jobs are important in representing large commercial computersystems. Typically, an exact analytical solution of such...
详细信息
ISBN:
(纸本)0897911415
Queueing network models of multiprogrammingsystems with memory constraints and multiple classes of jobs are important in representing large commercial computersystems. Typically, an exact analytical solution of such models is unavailable. Recently, a computationally efficient iterative technique has been proposed by A. Brandwajn, E. Lazowska and J. Zahorjan for models of systems in which each job is subject to a separate memory constraint, i. e. , has its own memory domain. It is not unusual, however, to have several jobs of different classes share a single memory 'domain' (e. g. , IBM's Information Management System). The authors present a simple approximate solution to the shared domain problem. The approach is inspired by the recently proposed technique which is complemented by a few approximations to preserve the conceptual simplicity and computational efficiency of this technique. The accuracy of the results is generally in fair agreement with simulation.
With the advent of Ada, it is possible to produce powerful generic packages that solve a broad class of problems. A common example that illustrates the power and utility of generics in Ada is a generic sorting package...
详细信息
With the advent of Ada, it is possible to produce powerful generic packages that solve a broad class of problems. A common example that illustrates the power and utility of generics in Ada is a generic sorting package. With such a package one can sort objects of almost any type providing that a procedure is supplied by the client program that defines the meaning of greater than, with respect to the objects that are being sorted. Modula-2 does not directly support generics. It is nevertheless possible to simulate the function of generics by using procedure parameters and open array parameters of words. A generic sorting module in Modula-2 is presented and illustrated. Two sorting algorithms are used;a simple exchange sort and the powerful, efficient heap sort. It would be easy to extend the power of the module with additional generic sorting procedures.
It has been claimed that the dispatching policy whereby higher preemptive priority is given to a more I/O-bound job at the CPU achieves high throughput of multiprogrammed computersystems. However the policy does not ...
详细信息
It has been claimed that the dispatching policy whereby higher preemptive priority is given to a more I/O-bound job at the CPU achieves high throughput of multiprogrammed computersystems. However the policy does not reflect the difference in how important the quick response is to each job. In this paper, first, a performance measure, called production value of the system, is introduced as an indicator of overall responsiveness which reflects the above difference. Secondly, a simple dispatching policy, which at the CPU is a generalization of the above policy, is proposed which is intended to optimize the performance measure at each time interval. The proposed policy is evaluated by means of queueing analysis and simulation.
This article describes an automatic display on a terminal screen of selected references to a labeled program or document element while indicating the selections in an index list displayed on the screen at the same tim...
详细信息
This article describes an automatic display on a terminal screen of selected references to a labeled program or document element while indicating the selections in an index list displayed on the screen at the same time. The index list may be a program cross-reference list or any other index of references within the program or document being studied. The referenced element can be a location in a program, a data element, or any other labeled component of the program or document. The display of a reference to an element consists of a display of the area of the program in which the element reference occurs. This area depends upon the form of representation used for the program, e. g. , one or more lines of a source code listing or one or more blocks of a flowchart centered about the location of the referenced element. One or more references can be selected and displayed at the same time.
We consider concrete algorithms for the realization of conditional operators with concurrent homogeneous operations in different branches of the multilevel conditional operator for SIMD computers.
We consider concrete algorithms for the realization of conditional operators with concurrent homogeneous operations in different branches of the multilevel conditional operator for SIMD computers.
A problem faced which arises in many computing applications is that of sorting the entries in a list into ascending sequence of their values. Efficient algorithms for sorting long lists usually require recursive techn...
详细信息
A problem faced which arises in many computing applications is that of sorting the entries in a list into ascending sequence of their values. Efficient algorithms for sorting long lists usually require recursive techniques and are not straightforward to program in a non-recursive language such as FORTRAN 77. This paper describes three FORTRAN 77 implementations of the insertion sort, quicksort and modified quicksort algorithms. The first algorithm is suitable for short lists while the second two are suitable for longer lists. A detailed description of the removal of recursion from the quicksort algorithm by the use of stacks is presented. Some timing statistics are given to illustrate the utility of the subroutines presented.
Recently, IBM has announced a program named Host Displaywriter Document Interchange (HDDI). HDDI allows a Displaywriter user to invoke two new functions. The first function transfers documents back and forth between t...
详细信息
Recently, IBM has announced a program named Host Displaywriter Document Interchange (HDDI). HDDI allows a Displaywriter user to invoke two new functions. The first function transfers documents back and forth between the Displaywriter's diskette and an attached VM/370 system's disk. The second function transforms documents back and forth between the Displaywriter's editable document form called OIIA L3 DCA and an editable document form used on VM/370, called SCRIPT/VS.
作者：
Wegner, Lutz M.Univ Karlsruhe
Inst fuer Angewandte Informatik und Formale Beschreibungsverfahren Karlsruhe Univ Karlsruhe Inst fuer Angewandte Informatik und Formale Beschreibungsverfahren Karlsruhe West
A new algorithm based on Quicksort for sorting in place a distributed file in a message switching network is presented. The algorithmic aspects of the method are discussed and it is compared to other possible approach...
详细信息
A new algorithm based on Quicksort for sorting in place a distributed file in a message switching network is presented. The algorithmic aspects of the method are discussed and it is compared to other possible approaches. The sort is analyzed and is shown to require, on the average, O(M multiplied by (times) log(M)) messages and a total data traffic of O(N multiplied by (times) log(M)) records for a file of size N fragmented over M stations. Arguments for the performance one may expect in practice, where N is much larger than M, are given. They are supported by results from a sequential simulation.
This paper discusses the relationship between machine organization and control program design in high-end commercial computersystems. The criterion is cost/performance, subject to achieving an acceptable performance ...
详细信息
This paper discusses the relationship between machine organization and control program design in high-end commercial computersystems. The criterion is cost/performance, subject to achieving an acceptable performance level. A brief discussion of the environment expected for the design and operation of high-end commercial computersystems is outlined, followed by a discussion of machine organization techniques which are classified and reviewed to permit a qualitative evaluation of the degree to which control program intent is exploited in machine organization. The thesis is developed next, using a hierarchical model which illustrates the contention that architecture has acted as a barrier to communication between the control program and machine organization. Examples of techniques that exploit knowledge of the intent of the control program and comments on the methodology that might be used to investigate such techniques follow. Directions for further research are then proposed.
When writing a Basic program that involves sorting of names or numbers, selection of the sorting method can be important to program speed. It is shown how sorting methods work, so informed selections of when and why o...
详细信息
When writing a Basic program that involves sorting of names or numbers, selection of the sorting method can be important to program speed. It is shown how sorting methods work, so informed selections of when and why one method should be chosen over another can be made. Some good textbooks are recommended to learn about sorting algorithms. A program is introduced to permit clearly the visualization of three common sorting methods on the screen. It dramatically shortens the time needed to understand why one method beats another. In operation, it builds a disordered array of ASCII characters, displays it and then proceeds to illustrate a step-by-step sort by the methods known as bubble sort, quicksort, and fastsort. As it goes, it displays a report on how many comparisons and swaps were necessary to achieve order.
暂无评论