Performance analysis of client-server software systems through bounds on task throughputs is presented. The upper and lower bounds (performance guarantee) are both independent of any assumptions about the stochastic behavior of the client tasks, and make only weak assumptions on the server. The analytic expressions for the bounds however, are not in closed form. A new technique based on interval arithmetic is developed to compute numerical values for the bounds.

To use queuing theory to analyze real systems such as computer communications networks, one makes assumptions that are, strictly speaking, untrue. The authors provide an exact analysis for cases with greatly relaxed assumptions. Service times can have general increasing failure rate distributions, different by class even at FIFO nodes. Routing can be arbitrary, including dependencies along the route, provided the number of visits to each node is a random variable. Only the mean service time and mean visit rates at nodes need be specified. A lower throughput bound is found which gives a minimum guaranteed throughput for each class; together with the familiar multiclass asymptotic upper bounds they give a convex feasible region in a multidimensional throughput space. A detailed analysis is given for systems with FIFO and infinite-server nodes, and the extension to processor-sharing nodes is described. The results can be reinterpreted as a set of bounds on the separate throughputs. This is equivalent to a circumscribed rectangular region called the robust box bounds.< >

The role of broker in client-server systems is to accommodate flexible, open, heterogeneous system design and to facilitate fault tolerance and improved performance through load balancing. Some architectural design decisions, such as broker distribution and server replication, strongly affect the system performance. The paper presents alternative client-broker-server architectures and compares their performance by using a combination of measurements and simulation. Broker distribution is found to reduce some of the overheads due to the introduction of brokers into client-server systems, and results in a system that has performance characteristics comparable to that of a pure client-server system. It is shown that in a client-server-broker system employing stateful server replication and using the unicast group communication mechanism for server-state consistency, there is a critical message size beyond which the system incurs degradation in performance.

Software bottlenecks are performance constraints caused by slow execution of a software task, in typical client-server systems a client task must wait in a blocked state for the server task to respond to its requests, so a saturated server will slow down all its clients. A rendezvous network generalizes this relationship to multiple layers of servers with send-and-wait interactions (rendezvous), a two-phase model of task behavior, and to a unified model for hardware and software contention. Software bottlenecks have different symptoms, different behavior when the system is altered, and a different cure from the conventional bottlenecks seen in queueing network models of computer systems, caused by hardware limits. The differences are due to the "push-back" effect of the rendezvous, which spreads the saturation of a server to its clients. The paper describes software bottlenecks by examples, gives a definition, shows how they can be located and alleviated, and gives a method for estimating the performance benefit to be obtained. Ultimately, if all the software bottlenecks can be removed, the performance limit will be due to a conventional hardware bottleneck.< >

Distributed or parallel software with synchronous communication via rendezvous is found in client-server systems and in proposed open distributed systems, in implementation environments such as Ada, V, remote procedure call systems, in transputer systems, and in specification techniques such as CSP, CCS and LOTOS. The delays induced by rendezvous can cause serious performance problems, which are not easy to estimate using conventional models which focus on hardware contention, or on a restricted view of the parallelism which ignores implementation constraints. Stochastic rendezvous networks are queueing networks of a new type which have been proposed as a modelling framework for these systems. They incorporate the two key phenomena of included service and a second phase of service. This paper extends the model to also incorporate different services or entries associated with each task. Approximations to arrival-instant probabilities are employed with a mean-value analysis framework, to give approximate performance estimates. The method has been applied to moderately large industrial software systems

Synchronous message-passing communication, or rendezvous, occurring between software tasks can have a significant effect on system performance. The rendezvous style of communication is coming into wider use in programming languages and operating systems for parallel and distributed environments. Understanding the performance implications of this style of inter-task communication is becoming a matter of practical importance. The dual nature of a task which acts both like a customer as well as a server, makes the performance analysis of rendezvous-based multitasking systems quite different from the analysis of the other queueing systems with known results. This research focuses on rendezvous-based systems in which the execution behavior of the software has a nondeterministic component of a very general nature which may for example be the manifestation of a data dependent behavior. Based on a model called the Stochastic Rendezvous Network the computation of bounds on task throughputs for multitasking systems characterized by rendezvous style communication is presented. Although the behavior of tasks is called stochastic, it is very general and the results are valid for general distributions of computation times and the number of messages generated by tasks. The inter-relationship among task throughputs, however, makes it difficult to extract the bounds in closed analytic form. The notion of a feasible throughput region which encloses the set of feasible tasks throughputs and captures the inter-relationship among the behavior of tasks is introduced. Variations of this basic bounding approach that are useful in the context of different types of multitasking systems are considered in the article. For example, a novel technique based on interval arithmetic is proposed for the computation of numerical values for the bounds. The applicability of the bounds and their tightness are analyzed through case studies. Issues such as the inter-relationship between the software architecture and system performance, and the effect of processor contention on the performance bounds are discussed.

Stochastic rendezvous networks (SRVN) are models for the throughput of distributed programs executed concurrently and synchronously on distributed computing nodes. Server tasks respond to requests from user tasks and execute in a two-phase pattern (in-rendezvous phase, then postrendezvous phase), and may themselves act as user tasks to further servers, to any depth. The authors extend the model to permit the execution time and the occurrence of nested rendezvous calls to depend on the identity of the calling user task. An iterative approximation method is developed, and compared to exact throughput calculations found with timed Petri nets. The approximation has errors of a few percent in most cases, and executes very much faster than the Petri-net calculation.< >

Multi-tasking multiprocessor systems frequently use rendezvous for intertask communication , and Stochastic Rendezvous Networks are a form of performance model for such systems. Tasks are frequently given diierent priorities to adjust the system performance. This paper considers a pre-emptive-resume priority discipline for executing tasks sharing each processor. An iterative algorithm is proposed, based on the well-known \MVA Priority Approximation". Throughput errors below 10% are found in many cases. The worst accuracy (as usual) is for low-priority tasks on highly-utilized processors.

In this correspondence we complement, by means of an analytical model, an earlier simulation study on tradeoff between delay and TASI advantage in a packetized speech multiplexer.

In this paper, a variable-frame hybrid multiplexer with a min-max constraint on frame length is proposed. Using a single server queueing model, it is shown that the performance of the variableframe multiplexer with respect to channel utilization, blocking, and delay is better than its fixed frame counterpart. Some of the results are compared to those from earlier studies.

When terminals (data and/or voice), connected to an asynchronous time division multiplexer (ATDM), are buffered and operate at low speed, the arrivals at the multiplexer are correlated. In this paper, we consider the finite buffer behavior of an ATDM with such terminals. An earlier study by Rudin [6] is a special case of this approach. The treatment here is also an improvement over the analysis by Birdsall et al. [1] of an ATDM with speech sources and is suitable for packet voice applications.

The dynamics of slowly spinning axisymmetric s a t e l l i t e s under the influence of gravity gradient torque i s investigated using a n a l y t i c a l and. numerical techniques. P a r t i c u l a r emphasis i s on motion near the equilibrium p o s i t i o n i n which the spin axis i s normal to the o r b i t a l plane. The problem i s studied i n increasing orders of d i f f i c u l t y . Phase I deals with the response and s t a b i l i t y of a s i m p l i f i e d model free to l i b r a t e i n r o l l while the more general problem i s treated i n Phase I I . Phase I serves as a proving ground fo r techniques to be used i n subsequent analysis. A closed form solution i s obtained i n terms of e l l i p t i c functions for the autonomous case. In general, f o r non-circular o r b i t s , motion i n the large i s studied using the concept of the invariant solution surface. These surfaces, obtained numerically, reveal the nature of motion i n the large i n terms of the dominant periodic solutions and allow one to determine the l i m i t s of o s c i l l a t o r y motion i n terms of the state parameters. Floquet theory i s employed i n conjunct i o n with numerical solutions of the l i n e a r i z e d equations of motion to study s t a b i l i t y i n the small. This technique i s extended to assess the v a r i a t i o n a l s t a b i l i t y of the dominant periodic motions i n the large. Phase II investigates a more general model with three degrees of freedom i n attitude motion. The presence of an ignorable coordinate gives a fourth order, non-autonomous system for an e l l i p t i c t r a j e c t o r y . Motion i n the small i s studied extensiveI l l l y , again using Floquet theory, and s t a b i l i t y charts suitable for design purposes are presented. The invariant surface concept i s successfully extended to the study of the autonomous case i n the large. Methods are developed for determining the maximum response to a given disturbance r e s u l t i n g i n a set of charts which are useful i n assessing the e f f e c t s of non-lineari t i e s and the v a l i d i t y of the analysis i n the small. Procedures are explained for determining periodic solutions of the problem, as well as t h e i r s t a b i l i t y , for a r b i t r a r y e c c e n t r i c i t y . The analysis suggests the p o s s i b i l i t y of attitude i n s t a b i l i t y during spin-up operations. I t i s shown that stable motion can be established by providing either a p o s i t i v e or negative spin to the s a t e l l i t e with the former preferrable. Given s u f f i c i e n t spin any configuration, even those with an adverse gravity gradient e f f e c t , can be s t a b i l i z e d . E c c e n t r i c i t y affects the attitude motion of a s a t e l l i t e adversely as regions of unstable motion increase i n size and number with i t . TABLE OF CONTENTS Chapter Page

The paper emphasizes the importance of periodic solutions in the dynamic stability study of an axi-symmetric satellite in presence of gravity gradient torques. Initial conditions for periodic solutions are presented over a range of system parameters for motion in circular and elliptic orbits. The variational stability of periodic solutions is examined using an extension of the Floquet theory to a fourth order system.

Viscous flows in cavities having the cross-sectional shape of circular angle sectors are considered. The complete range of angles from π/12 to π is investigated numerically, using a new approximation method. The occurrence of higher-order vortices in the corner region is verified and numerical values for both the stream function and location of vortex centers are obtained. Experimental data were taken for sectors with three different opening angles.

AbstractThe concentration of solute in a film flowing down an inclined plane has been obtained by solving the complete diffusion equation, including the diffusion effect in the direction of flow. The formulation leads to the necessity of solving non‐orthogonal characteristic value problem, treatment of which is, however, quite straightforward.Though the motivation for the study was the diffusion of oxygen to plasma or blood films, solutions are presented for the complete range of fluid properties, due to the great significance of this mode of flow for wider application in chemical engineering.

Attitude dynamics of a spinning, axi-symmetric, rigid body undergoing central force motion has received considerable attention in recent times. Thomson et al studied the problem where the satellite was restricted to follow a circular trajectory. This stipulation reduced the system to an autonomous form which was then treated by linearised or Liapounov type of analysis. For the satellite in an elliptic orbit, Kane and Barba presented numerical solution to the linearised equations of motion. Recently Wallace and Meirovitch investigated the same problem by performing asymptotic analysis on linear and low order non-linear systems.

SummaryThe attitude dynamics of a spinning axi-symmetric satellite, free to librate in roll, is studied using the WKBJ method. Applicability of the method to the linearised equation of motion is first verified. To facilitate the evaluation of the WKBJ solution and to make it more amenable to the phase space analysis, second order terms in eccentricity are neglected. The form of the solution suggests the presence of phase and amplitude modulation effects for non-circular orbits. The comparison between the WKBJ analysis and ‘exact’ numerical solutions showed that the approximate analysis predicts the period and amplitude of motion with an accuracy adequate for the preliminary design purposes.