Two Tools for Interprocess Communication in Distributed Data-Flow Systems

Abstract is not available.

DFL: A Data Flow Language

Many novel computer architectures like array and multiprocessors which achieve high performance through the use of concurrency exploit variations of the von Neumann model of computation. The effective utilization of the machines makes special demands on programmers and their programming languages, such as the structuring of data into vectors or the partitioning of programs into concurrent processes. In comparison, the data flow model of computation demands only that the principle of structured programming be followed. A data flow program, often represented as a data flow graph, is a program that expresses a computation by indicating the data dependencies among operators. A data flow computer is a machine designed to take advantage of concurrency in data flow graphs by executing data independent operations in parallel. In this paper, we discuss the design of a high level language (DFL: Data Flow Language) suitable for data flow computers. Some sample procedures in DFL are presented. The implementation aspects have not been discussed in detail since there are no new problems encountered. The language DFL embodies the concepts of functional programming, but in appearance closely resembles Pascal. The language is a better vehicle than the data flow graph for expressing a parallel algorithm. The compiler has been implemented on a DEC 1090 system in Pascal.

Improving Flow-Insensitive Solutions for Non-Separable Dataflow Problems

Flow-insensitive solutions to dataflow problems have been known to be highly scalable; however also hugely imprecise. For non-separable dataflow problems this solution is further degraded due to spurious facts generated as a result of dependence among the dataflow facts. We propose an improvement to the standard flow-insensitive analysis by creating a generalized version of the dominator relation that reduces the number of spurious facts generated. In addition, the solution obtained contains extra information to facilitate the extraction of a better solution at any program point, very close to the flow-sensitive solution. To improve the solution further, we propose the use of an intra-block variable renaming scheme. We illustrate these concepts using two classic non-separable dataflow problems --- points-to analysis and constant propagation.

Real-fluid effects in flow cavitation

The possible role of reaj fluid effects in two aspects of flow cavitation namely inception and separation is discussed. This is primarily qualitative in the case of inception whereas some quantitative results are presented in the case of separation. Existing evidence clearly indicates that in particular viscous effects can play a significant role in determining the conditions for cavitation inception and in determining the location of cavitation separation from smooth bodies.

On the unsteady flow of a fairly concentrated suspension due to contracting or expanding walls of a tube

Consummating our earlier work [1], the unsteady flow of a fairly concentrated suspension due to a single contraction or expansion of the walls of a tube is studied. A comparison of the results obtained by using two different formulae for the additional drag terms in the governing equations has been made. A region of circulation in the flow field is observed when the volume fraction Z greater-or-equal, slanted 0.3, the Schmidt number Sc < 1 and the density ratio (density of the particulate phase/density of the fluid phase) > 1.

R-parameter: A local truncation error based adaptive framework for finite volume compressible flow solvers

A residual-based strategy to estimate the local truncation error in a finite volume framework for steady compressible flows is proposed. This estimator, referred to as the -parameter, is derived from the imbalance arising from the use of an exact operator on the numerical solution for conservation laws. The behaviour of the residual estimator for linear and non-linear hyperbolic problems is systematically analysed. The relationship of the residual to the global error is also studied. The -parameter is used to derive a target length scale and consequently devise a suitable criterion for refinement/derefinement. This strategy, devoid of any user-defined parameters, is validated using two standard test cases involving smooth flows. A hybrid adaptive strategy based on both the error indicators and the -parameter, for flows involving shocks is also developed. Numerical studies on several compressible flow cases show that the adaptive algorithm performs excellently well in both two and three dimensions.

Pulsatile blood flow in a rigid pulmonary alveolar sheet with porous walls

This paper deals with the pulsatile blood flow in the lung alveolar sheets by idealizing each of them as a channel covered by porous media. As the blood flow in the lung is of low Reynolds number, a creeping flow is assumed in the channel. The analytical and numerical results for the velocity and pressure distribution in the porous medium are presented. The effect of an imposed slip condition is also studied. Comparisons with the corresponding results for the steady-state case are made at the end.

Flow of a micropolar fluid in a tube with sinusoidal deformation

The flow of a micropolar fluid for sinusoidally deforming boundaries is discussed in detail. The velocity and microrotation fields and the streamfunction are determined and plotted under the assumption of small deformations.

A Method of Characteristics for Solving Two-dimensional Unsteady Flow Problems

The aim of this investigation is to evolve a method of solving two-dimensional unsteady flow problems by the method of characteristics. This involves the reduction of the given system of equations to an equivalent system where only interior derivatives occur on a characteristic surface. From this system, four special bicharacteristic directional derivatives are chosen. A finite difference scheme is prescribed for solving the equations. General rectangular lattices are also considered. As an example, we investigate the propagation of an initial pressure distribution in a medium at rest.

Pulsatile flow in tubes of varying cross-sections

The pulsatile flow of an incompressible viscous fluid in a cylindrical tube of varying cross section is investigated for small Reynolds numbers. The solutions consist of a stedy and an oscillatory part. The shear stress distribution on the wall is evaluated and discussed in detail for special geometries like tapered tubes, locally constricted tubes and peristaltic tubes. The existence of separation in the flow field is noticed.

Parametric Study of Flood Wave Propagation

A parametric study of the flood wave propagation problem is made, based on numerical solution of the nondimensionalized unsteady flow equations of open channels. The propagation of a sinusoidal flood wave in a prismatic channel is studied for uniform initial flow. The governing parameters (initial uniform flow Froude number, wave amplitude, wave duration, channel width parameter and side slope) are varied over a wide range. In all, 49 cases are studied. Effects of these governing parameters on the subsidence of stage and discharge and the speed of the wave peak are described in detail. The relative wave amplitude is found to vary linearly with F0, the initial uniform flow froude number, for lower F0 values. Wave duration has a very pronounced effect on subsidence with greater subsidence at lower wave duration values.

Flow-graph techniques for epicyclic gear train analysis

Flow-graph techniques are applied in this article for the analysis of an epicyclic gear train. A gear system based on this is designed and constructed for use in Numerical Control Systems.

A parametric differentiation version with finite-difference scheme applicable to a class of problems in boundary layer flow with massive blowing

A numerical procedure, based on the parametric differentiation and implicit finite difference scheme, has been developed for a class of problems in the boundary-layer theory for saddle-point regions. Here, the results are presented for the case of a three-dimensional stagnation-point flow with massive blowing. The method compares very well with other methods for particular cases (zero or small mass blowing). Results emphasize that the present numerical procedure is well suited for the solution of saddle-point flows with massive blowing, which could not be solved by other methods.

Engine wear and used oil analysis - comparative study in high speed diesel engines

The operational life and reliability of I.C. engines are limited to a certain extent by the break down of the engine components due to wear. It is advantageous to know the condition of an engine and its components without disassembling for detailed measurements. This paper describes the possibility of employing chemical analysis of the used crank case oil to predict the wear of engine components. It is concluded that the acidity and carbon contents of the crank case oil play a significant role in assessing the wear of copper-lead bearings used for the big end of the connecting rod.