Modelling and analysis of the variance in parallelism in parallel computations

In this paper, we introduce an analytical technique based on queueing networks and Petri nets for making a performance analysis of dataflow computations when executed on the Manchester machine. This technique is also applicable for the analysis of parallel computations on multiprocessors. We characterize the parallelism in dataflow computations through a four-parameter characterization, namely, the minimum parallelism, the maximum parallelism, the average parallelism and the variance in parallelism. We observe through detailed investigation of our analytical models that the average parallelism is a good characterization of the dataflow computations only as long as the variance in parallelism is small. However, significant difference in performance measures will result when the variance in parallelism is comparable to or higher than the average parallelism.

Minimization of grid current distortion in parallel-connected converters through carrier interleaving

Identical parallel-connected converters with unequal load sharing have unequal terminal voltages. The difference in terminal voltages is more pronounced in case of back-to-back connected converters, operated in power-circulation mode for the purpose of endurance tests. In this paper, a synchronous reference frame based analysis is presented to estimate the grid current distortion in interleaved, grid-connected converters with unequal terminal voltages. Influence of carrier interleaving angle on rms grid current ripple is studied theoretically as well as experimentally. Optimum interleaving angle to minimize the rms grid current ripple is investigated for different applications of parallel converters. The applications include unity power factor rectifiers, inverters for renewable energy sources, reactive power compensators, and circulating-power test set-up used for thermal testing of high-power converters. Optimum interleaving angle is shown to be a strong function of the average of the modulation indices of the two converters, irrespective of the application. The findings are verified experimentally on two parallel-connected converters, circulating reactive power of up to 150 kVA between them.

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

Thermal Degradation Processes in Poly(xylylene sulfides) Investigated by Comparative Direct Pyrolysis MS and Flash Pyrolysis GC/MS Experiments

The thermal degradation processes of two sulfur polymers, poly(xylylene sulfide) (PXM) and poly(xylylene disulfide) (PXD), were investigated in parallel by direct pyrolysis mass spectrometry (DPMS) and flash pyrolysis GC/MS (Py-GC/MS). Thermogravimetric data showed that these polymers decompose with two separate steps in the temperature ranges of 250-280 and 600-650 degrees C, leaving a high amount of residue (about 50% at 800 degrees C). The pyrolysis products detected by DPMS in the first degradation step of PXM and PXD were terminated by three types of end groups, -CH3, -CH2SH, and -CH=S, originating from thermal cleavage reactions involving a series of homolytic chain scissions followed by hydrogen transfer reactions, generating several oligomers containing some intact xylylene sulfide repeating units. The presence of pyrolysis compounds containing some stilbene-like units in the first degradation step has also been observed. Their formation has been accounted for with a parallel cleavage involving the elimination of H2S from the PXM main chains. These unsaturated units can undergo cross-linking at higher temperatures, producing the high amount of char residue observed. The thermal degradation compounds detected by DPMS in the second decomposition step at about 600-650 degrees C were constituted of condensed aromatic molecules containing dihydrofenanthrene and fenanthrene units. These compounds might be generated from the polymer chains containing stilbene units, by isomerization and dehydrogenation reactions. The pyrolysis products obtained in the Py-GC/MS of PXM and PXD at 610 degrees C are almost identical. The relative abundance in the pyrolysate and the spectral properties of the main pyrolysis products were found to be in generally good agreement with those obtained by DPMS. Polycyclic aromatic hydrocarbons (PAHs) were also detected by Py-GC/MS but in minor amounts with respect to DPMS. This apparent discrepancy was due to the simultaneous detection of PAHs together with all pyrolysis products in the Py-GC/MS, whereas in DPMS they were detected in the second thermal degradation step without the greatest part of pyrolysis compounds generated in the first degradation step. The results obtained by DPMS and PSI-GC/MS experiments showed complementary data for the degradation of PXM and PXD and, therefore, allowed the unequivocal formulation of the thermal degradation mechanism for these sulfur-containing polymers.

A new parallel algorithm for parsing arithmetic infix expressions

A new parallel algorithm for transforming an arithmetic infix expression into a par se tree is presented. The technique is based on a result due to Fischer (1980) which enables the construction of the parse tree, by appropriately scanning the vector of precedence values associated with the elements of the expression. The algorithm presented here is suitable for execution on a shared memory model of an SIMD machine with no read/write conflicts permitted. It uses O(n) processors and has a time complexity of O(log2n) where n is the expression length. Parallel algorithms for generating code for an SIMD machine are also presented.

A parallel wilf algorithm for complex zeros of a polynomial

A global recursive bisection algorithm is described for computing the complex zeros of a polynomial. It has complexityO(n 3 p) wheren is the degree of the polynomial andp the bit precision requirement. Ifn processors are available, it can be realized in parallel with complexityO(n 2 p); also it can be implemented using exact arithmetic. A combined Wilf-Hansen algorithm is suggested for reduction in complexity.

Mechanism of adsorption effects on electrode kinetics: Part I: Selection rules in the case of low coverage by molecular surfactants

It is shown that the effect of adsorption of inert molecules on electrode reaction rates is completely accounted for, by introducing into the rate equation, adsorption-induced changes in both the effective electrode area as well as in the electrostatic potential at the reaction site with an additional term for the noncoulombic interaction between the reactant and the adsorbate. The electrostatic potential at the reaction site due to the adsorbed layer is calculated using a model of discretely-distributed molecules in parallel orientation when adsorbed on the electrode with an allowance for thermal agitation. The resulting expression, which is valid for the limiting case of low coverages, is used to predict the types of molecular surfactants that are most likely to be useful for acceleration and inhibition of electrode reactions.

Amine levels in Lathyrus sativus seedlings during development

In growing Lathyrus sativus seedlings, the levels of DNA, RNA and protein markedly decreased in the cotyledons and progressively increased in the embryo-axis. In cotyledons, spermidine and spermine contents were substantially reduced while those of agmatine and putrescine were sharply increased. By contrast the embryo-axis progressively accumulated relatively larger amounts of agmatine, homoagmatine. putrescine, cadaverine, spermidine and spermine in parallel with similar changes in its DNA, RNA and protein content. While the cotyledons contained ca 50% of the total agmatine and putrescine present in the plant embryo by day 10, the embryo-axis, though representing less than 20% of the dry wt, contained 90 and 75% of total cadaverine and homoagmatine respectively of the seedlings. Spermidine and spermine levels of this tissue were also comparatively higher, being of the order of 80 and 50% respectively of the total. The root and shoot portions of the embryo-axis also exhibited a similar relationship between changes in DNA, RNA and protein and all the above amines during development. However, the polyamine content of the shoots was relatively higher than those of the roots during the growth period.

Optic phonon-magnetoplasmon surface oscillations in polar semiconductors

An interface between two polar semiconductors in parallel magnetic field geometry can support at most four types of surface oscillations; the actual number (less-than-or-equals, slant4), however, depends on the strength of the magnetic field. The interface effects on these relevant ranges of magnetic field are analysed in detail.

Expression of hyoscyamine 6 beta-hydroxylase in the root pericycle cells and accumulation of its product scopolamine in leaf and stem tissues of Datura metel L.

Hyoscyamine 60-hydroxylase (H6H: EC 1.14.11.11), a key enzyme at the terminal step of tropane alkaloid biosynthesis, converts hyoscyamine to scopolamine. The accumulation of scopolamine in different organs, in particular the aerial parts for storage, is subject to the expression of hyoscyamine 6-phydroxylase as well as its transport from the site of synthesis. To understand the molecular basis of this regulation, we have analyzed, in parallel, the relative levels of hyoscyamine and scopolamine, and the accumulation of H6H (both protein and transcript) in leaves, stems and roots of D. metel. The root, stem and leaf tissues all contain about 0.51-0.65 mg g(-1) dry weight of scopolamine. Hyoscyamine content was extremely low in leaf and stem tissues and was about 0.28 mg g(-1) dry weight in the root tissue. H6H protein and its transcript were found only in roots but not in the aerial parts viz. stems and leaves. The immunolocalization studies performed on leaf, stem, root as well as hairy root tissues showed that H6H was present only in the pericycle cells of young lateral and hairy roots. These studies suggest that the conversion of hyoscyamine to scopolamine takes place in the root pericycle cells, and the alkaloid biosynthesized in the roots gets translocated to the aerial parts in D. metel. (C) 2009 Elsevier Ireland Ltd. All rights reserved.

Dipole patterns in the structures of some ferroelectrics and antiferroelectrics

The dipole patterns in the ferroelectric and antiferroelectric structures are drawn according to experimentally determined symmetry changes in the ferroelectrics and antiferroelectrics. For the ferroelectrics the dipoles of the unit cells for one domain are oriented in parallel and the directions of the polarisation in the adjacent domains are at definite angles to each other. It is assumed for the antiferroelectrics, that the superstructural unit cell is formed by the adjacent cells of the paraelectrical modification; the subcells having the antiparallel directions of the polarisation. It is these superstructural cells of the antiferroelectrics that are determined during the experimental investigations of the antiferroelectrics. The superstructural cells of the adjacent domains are different. In one case, the difference is that in the adjacent domains, the directions of the polarisation in the subcells form an angle (e.g., in PbZrO3). In other cases the superstructural cells have not only different directions of the polarisation in the subcells but different signs of the enantiomorphism (e.g., NH4H2PO4). In the third case, the only difference is that the superstructural unit cells in the adjacent domains are turned by an angle to each other round the direction of the subcell polarisation [e.g., (NH4)2H3IO6], etc.

Singularity and controllability analysis of parallel manipulators and closed-loop mechanisms

This paper presents a study of kinematic and force singularities in parallel manipulators and closed-loop mechanisms and their relationship to accessibility and controllability of such manipulators and closed-loop mechanisms, Parallel manipulators and closed-loop mechanisms are classified according to their degrees of freedom, number of output Cartesian variables used to describe their motion and the number of actuated joint inputs. The singularities in the workspace are obtained by considering the force transformation matrix which maps the forces and torques in joint space to output forces and torques ill Cartesian space. The regions in the workspace which violate the small time local controllability (STLC) and small time local accessibility (STLA) condition are obtained by deriving the equations of motion in terms of Cartesian variables and by using techniques from Lie algebra.We show that for fully actuated manipulators when the number ofactuated joint inputs is equal to the number of output Cartesian variables, and the force transformation matrix loses rank, the parallel manipulator does not meet the STLC requirement. For the case where the number of joint inputs is less than the number of output Cartesian variables, if the constraint forces and torques (represented by the Lagrange multipliers) become infinite, the force transformation matrix loses rank. Finally, we show that the singular and non-STLC regions in the workspace of a parallel manipulator and closed-loop mechanism can be reduced by adding redundant joint actuators and links. The results are illustrated with the help of numerical examples where we plot the singular and non-STLC/non-STLA regions of parallel manipulators and closed-loop mechanisms belonging to the above mentioned classes. (C) 2000 Elsevier Science Ltd. All rights reserved.

A parallel multistep predictor-corrector algorithm for solving ordinary differential equations

In this paper we give a generalized predictor-corrector algorithm for solving ordinary differential equations with specified initial values. The method uses multiple correction steps which can be carried out in parallel with a prediction step. The proposed method gives a larger stability interval compared to the existing parallel predictor-corrector methods. A method has been suggested to implement the algorithm in multiple processor systems with efficient utilization of all the processors.

A FIFO-based multicast network and its use in multicomputers

We present an implementation of a multicast network of processors. The processors are connected in a fully connected network and it is possible to broadcast data in a single instruction. The network works at the processor-memory speed and therefore provides a fast communication link among processors. A number of interesting architectures are possible using such a network. We show some of these architectures which have been implemented and are functional. We also show the system software calls which allow programming of these machines in parallel mode.

Power Flow Analysis in MTDC-AC System—A New Approach

This paper presents a new approach to the power flow analysis in steady state for multiterminal DC-AC systems. A flexible and practical choice of per unit system is used to formulate the DC network and converter equations. A converter is represented by Norton's equivalent of a current source in parallel with the commutation resistance. Unlike in previous literature, the DC network equations are used to derive the controller equations for the DC system using a subset of specifications. The specifications considered are current or power at all terminals except the slack terminal where the DC voltage is specified. The control equations are solved by Newton's method, using the current injections at the converter terminals as state variables. Further, a systematic approach to the handling of constraints is proposed by identifying the priorities in rescheduling of the specified variables. The methodology is illustrated by example of a 5 terminal DC system.