Exploiting Programmable Network Interfaces for Parallel Query Execution in Workstation Clusters

Workstation clusters equipped with high performance interconnect having programmable network processors facilitate interesting opportunities to enhance the performance of parallel application run on them. In this paper, we propose schemes where certain application level processing in parallel database query execution is performed on the network processor. We evaluate the performance of TPC-H queries executing on a high end cluster where all tuple processing is done on the host processor, using a timed Petri net model, and find that tuple processing costs on the host processor dominate the execution time. These results are validated using a small cluster. We therefore propose 4 schemes where certain tuple processing activity is offloaded to the network processor. The first 2 schemes offload the tuple splitting activity - computation to identify the node on which to process the tuples, resulting in an execution time speedup of 1.09 relative to the base scheme, but with I/O bus becoming the bottleneck resource. In the 3rd scheme in addition to offloading tuple processing activity, the disk and network interface are combined to avoid the I/O bus bottleneck, which results in speedups up to 1.16, but with high host processor utilization. Our 4th scheme where the network processor also performs apart of join operation along with the host processor, gives a speedup of 1.47 along with balanced system resource utilizations. Further we observe that the proposed schemes perform equally well even in a scaled architecture i.e., when the number of processors is increased from 2 to 64

Fast matrix transposition computer implementation

Eklundh's (1972) algorithm to transpose a large matrix stored on an external device such as a disc has been programmed and tested. A simple description of computer implementation is given in this note.

Two-dimensional phase transitions at electrochemical interfaces

A two-state Ising model has been applied to the two-dimensional condensation of tymine at the mercury-water interface. The model predicts a quadratic dependence of the transition potential on temperature and on the logarithm of the adsorbate concentration. Both predictions have been confirmed experimentally.

Salan: A microprocessor-based medium speed experimental local area network

In this paper the implementation and application of a microprocessor-based medium speed experimental local area network using a coaxial cable transmission medium are dealt with. A separate unidirectional control wire has been used in order to provide a collision-free and fair medium access arbitration. As an application of the network, the design of a packet voice communication system is discussed.

Computer simulation of polytypes

Polytypes have been simulated, treating them as analogues of a one-dimensional spin-half Ising chain with competing short-range and infinite-range interactions. Short-range interactions are treated as random variables to approximate conditions of growth from melt as well as from vapour. Besides ordered polytypes up to 12R, short stretches of long-period polytypes (up to 33R) have been observed. Such long-period sequences could be of significance in the context of Frank's theory of polytypism. The form of short-range interactions employed in the study has been justified by carrying out model potential calculations.

Development of a computer code to predict cooling of die and billet in forging

A hot billet in contact with relatively cold dies undergoes rapid cooling in the forging operation. This may give rise to unfilled cavities, poor surface finish and stalling of the press. A knowledge of billet-die temperatures as a function of time is therefore essential for process design. A computer code using finite difference method is written to estimate such temperature histories and validated by comparing the predicted cooling of an integral die-billet configuration with that obtained experimentally.

Computer-aided analysis of the structure of kinematic chains

Using the link-link incidence matrix to represent a simple-jointed kinematic chain algebraic procedures have been developed to determine its structural characteristics such as the type of freedom of the chain, the number of distinct mechanisms and driving mechanisms that can be derived from the chain. A computer program incorporating these graph theory based procedures has been applied successfully for the structural analysis of several typical chains.

Computer model for vedavati ground water basin. Part 1. Well field model

It is shown that a leaky aquifer model can be used for well field analysis in hard rock areas, treating the upper weathered and clayey layers as a composite unconfined aquitard overlying a deeper fractured aquifer. Two long-duration pump test studies are reported in granitic and schist regions in the Vedavati river basin. The validity of simplifications in the analytical solution is verified by finite difference computations.

Computer-model for vedavati groundwater basin .3. Irrigation potential

An estimate of the irrigation potential over and above the existing utilization was made based on the ground water potential in the Vedavati river basin. The estimate is based on assumed crops and cropping patterns as per existing practice in the various taluks of the basin. Irrigation potential was estimated talukwise based on the available ground water potential identified from the simulation study. It is estimated that 84,100 hectares of additional land can be brought under irrigation from ground water in the entire basin.

Complex carbohydrates: 2. The modes of binding of complex carbohydrates to concanavalin A - a computer modelling approach

The probable modes of binding of some complex carbohydrates, which have the trimannosidic core structure (Man3GlcNAc2), to concanavalin A (Con A) have been determined using a computer modelling technique. These studies show that Con a can bind to the terminal mannose residues of the trimannosidic core structure and to the internal mannosyl as well as to the terminal N-acetylglucosamine residues of the N-acetylglucosamine substituted trimannosidic core structure. The oligosaccharide with terminal mannose residues can bind in its minimum energy conformers, whereas the oligosaccharide with internal mannosyl and terminal N-acetylglucosamine residues can bind only in higher energy conformers. In addition the former oligosaccharide forms more hydrogen bonds with Con A than the latter. These results suggest that, for these oligosaccharides, the terminal mannose residue has a much higher probability of reaching the binding site than either the internal mannosyl or the terminal N-acetylglucosamine residues. The substitution of a bisecting N-acetylglucosamine residue on these oligosaccharides, affects significantly the accessibility of the residues which bind to Con A and thereby reduces their binding affinity. It thus seems that the binding affinity of an oligosaccharide to Con A depends not only on the number of sugar residues which possess free 3-, 4- and 6-hydroxyl groups but also on the accessibility of these sugar residues to Con A. This study also reveals that the sugar binding site of Con A is small and that the interactions between Con A and carbohydrates are extended slightly beyond the single sugar residue that is placed in the binding site.

Design and Performance Evaluation of EXMAN: An EXtended MANchester Data Flow Computer

Data flow computers are high-speed machines in which an instruction is executed as soon as all its operands are available. This paper describes the EXtended MANchester (EXMAN) data flow computer which incorporates three major extensions to the basic Manchester machine. As extensions we provide a multiple matching units scheme, an efficient, implementation of array data structure, and a facility to concurrently execute reentrant routines. A simulator for the EXMAN computer has been coded in the discrete event simulation language, SIMULA 67, on the DEC 1090 system. Performance analysis studies have been conducted on the simulated EXMAN computer to study the effectiveness of the proposed extensions. The performance experiments have been carried out using three sample problems: matrix multiplication, Bresenham's line drawing algorithm, and the polygon scan-conversion algorithm.

Linear description of electrical networks for computer analysis

A parentheses-free code is suggested for the description of two-terminal electrical networks for computer analysis.

Implementation of placement and routing algorithms for computer aided design of printed circuit boards

The Printed Circuit Board (PCB) layout design is one of the most important and time consuming phases during equipment design process in all electronic industries. This paper is concerned with the development and implementation of a computer aided PCB design package. A set of programs which operate on a description of the circuit supplied by the user in the form of a data file and subsequently design the layout of a double-sided PCB has been developed. The algorithms used for the design of the PCB optimise the board area and the length of copper tracks used for the interconnections. The output of the package is the layout drawing of the PCB, drawn on a CALCOMP hard copy plotter and a Tektronix 4012 storage graphics display terminal. The routing density (the board area required for one component) achieved by this package is typically 0.8 sq. inch per IC. The package is implemented on a DEC 1090 system in Pascal and FORTRAN and SIGN(1) graphics package is used for display generation.