METRICS

In the last two decades there has been special interest in queueing networks with a product form solution. These have been widely used as models of computer systems and communication networks. Two new computational algorithms for product form networks ...

The effect that parameter estimation errors have on performance in closed product form queueing networks is investigated. In particular, the effects of errors in the relative utilization estimates of the servers are analyzed. It is shown that in single ...

XRAY presents a global view of the performance of hardware and software components on multiple, distributed computers. The set of components chosen for measurement can be changed at any time throughout a network of systems, and can be selected to ...

This paper describes the design and application of a special purpose computer system. It was developed as an internal tool by a computer manufacturer, and has been used in solving a variety of measurement problems encountered in computer performance ...

A methodology for benchmarking dedicated, interactive systems has been developed at The MITRE Corporation. This methodology uses a synthetic program model of the application which runs on the proposed hardware/operating system configurations and is ...

This paper proposes a modelling methodology combining simulation and analysis for computer performance evaluation and prediction. The methodology is based on a special workload model that is suitable for the generation and description of dynamic program ...

Program behaviour characteristics were examined using data gathered from real program executions. Experiments were performed in a segmented virtual memory with a working set policy; the analyzing costs were kept low using an efficient data reduction ...

Prior studies of program restructuring to increase the degree of locality of a program in a paged virtual memory system were restricted to statically allocated codes only. This work develops a restructuring methodology for block structured languages ...

The near-complete decomposability of queueing network models of computer systems is generally supported by very large differences in the service rates of the servers. In this paper we show how such models may still be nearly completely decomposable if ...

This paper addresses three aspects related to the notion of exact equivalence in queueing models.

In many cases the parameters of a system equivalent to a given model involve only a small subset of conditional probabilities of the state of the original ...

Queueing networks have been used to model a large variety of complex systems. However, once a realistic model has been constructed it has generally been necessary to distort and modify it so that an analytic solution could be obtained. Unfortunately, ...

Equilibrium state distributions are determined for queues with load-dependent Poisson arrivals and service time distributions representable by Cox's generalized method of stages. The solution is obtained by identifying a birth-death process that has the ...

This paper extends a previous model for computer system configuration planning developed by the authors. The problem is to optimally select the CPU speed, the device capacities, and file assignments so as to maximize throughput subject to a fixed cost ...

This paper presents a model of an I/O subsystem in which devices can be accessed from multiple CPUs and/or via alternative channel and control unit paths. The model estimates access response times, given access rates for all CPU-device combinations. The ...

This paper presents several computer capacity planning case studies using a modeling tool, BEST/1, derived from the theory of queueing networks. All performance predictions were evaluated based on the selected service levels such as response times and ...

In this paper, we study the end-to-end control through virtual circuits in a computer network built following the X.25 Recommendations. We develop a mathematical model to obtain the maximum overallocation of node buffers, in order for the probability of ...

This paper describes an application of analytical modeling to the design and evaluation of a general purpose, packet-switched image processing system that will soon enter an implementation phase. A bottom-up modeling approach is used to evaluate such ...

Distribution of computing to achieve goals such as enhanced reliability depend on the use of decentralized software. Decentralization typically replaces a sequential process by a system of small, concurrent processes that interact frequently. The ...

Applicative programming Languages provide opportunities for parallel processing without requiring the programmer to be concerned with explicit synchronization of portions of the computation. We present a computational model of a multiprocessor which ...

This paper presents a method to estimate the performance of a feedback data flow processor using software simulation. A brief over view of a data flow language and a data flow processor along with the conceptual view of a software simulator are ...

Operational analysis replaces certain classical gueueing theory assumptions with the condition of “homogeneous service times.” In this paper, we show that the sample paths of an M/G/1 queueing system have this property with non-zero probability if and ...

Frequently used computer elements that can be written by at most one process at a time constitute important bottlenecks in multiprocessor system operation, particularly when such elements are accessible only serially. Hardware devices, data files, ...

We consider two simple models of overhead in batch computer systems and demand access communications systems. The first, termed “modified M/M/1/K, ” is an exponential, single-server queuing system with finite storage capacity, constant arrival rate, and ...

A single server queue with feedback and multiple customer classes is analyzed. Arrival processes are independent Poisson processes. Each round of service is exponentially distributed. After receiving a round of service, a customer may depart or rejoin ...

We consider a priority queueing system which consists of two queues sharing a processor and in which there is delayed feedback. Such a model arises from systems which employ a priority assignment scheme to achieve overload control. An analytic ...

This paper extends previous work on development of a methodology for the prediction of the performance of computer software systems from design level specifications and continuing through implementation. The effects of synchronized behavior, such as ...

Quantitative analysis of locking mechanisms and of their impact on the performance of transactionnal systems have yet received relatively little attention. Although numerous concurrency mechanisms have been proposed and implemented, there is an obvious ...

We consider the effect on system performance of the distribution of a data base in the form of multiple copies at distinct sites. The purpose of our analysis is to determine the gain in READ throughput that can be obtained in the presence of consistency ...

Policy function schedulers provide a flexible framework for implementing a wide range of different scheduling schemes. In such schedulers, the priority of a job at any instant in time is defined by the difference between the time it spent in the system ...

Real-time multiprogramming systems, in which a hardware processor is dynamically assigned to run multiple software processes each designed to control an important device (user), are considered. Each software process executes a task in response to a ...