Automatic code generation of overlapped communications in a parallelisation tool

This paper addresses the exploitation of overlapping communication with calculation within parallel FORTRAN 77 codes for computational fluid dynamics (CFD) and computational structured dynamics (CSD). The obvious objective is to overlap interprocessor communication with calculation on each processor in a distributed memory parallel system and so improve the efficiency of the parallel implementation. A general strategy for converting synchronous to overlapped communication is presented together with tools to enable its automatic implementation in FORTRAN 77 codes. This strategy is then implemented within the parallelisation toolkit, CAPTools, to facilitate the automatic generation of parallel code with overlapped communications. The success of these tools are demonstrated on two codes from the NAS-PAR and PERFECT benchmark suites. In each case, the tools produce parallel code with overlapped communications which is as good as that which could be generated manually. The parallel performance of the codes also improve in line with expectation.

Evaluation of loadsharing algorithms for heterogeneous distributed systems

The performance of loadsharing algorithms for heterogeneous distributed systems is investigated by simulation. The systems considered are networks of workstations (nodes) which differ in processing power. Two parameters are proposed for characterising system heterogeneity, namely the variance and skew of the distribution of processing power among the network nodes. A variety of networks are investigated, with the same number of nodes and total processing power, but with the processing power distributed differently among the nodes. Two loadsharing algorithms are evaluated, at overall system loadings of 50% and 90%, using job response time as the performance metric. Comparison is made with the ideal situation of ‘perfect sharing’, where it is assumed that the communication delays are zero and that complete knowledge is available about job lengths and the loading at the different nodes, so that an arriving job can be sent to the node where it will be completed in the shortest time. The algorithms studied are based on those already in use for homogeneous networks, but were adapted to take account of system heterogeneity. Both algorithms take into account the differences in the processing powers of the nodes in their location policies, but differ in the extent to which they ‘discriminate’ against the slower nodes. It is seen that the relative performance of the two is strongly influenced by the system utilisation and the distribution of processing power among the nodes.

Exact colorations of graphs and digraphs

A coloration is an exact regular coloration if whenever two vertices are colored the same they have identically colored neighborhoods. For example, if one of the two vertices that are colored the same is connected to three yellow vertices, two white and red, then the other vertex is as well. Exact regular colorations have been discussed informally in the social network literature. However they have been part of the mathematical literature for some time, though in a different format. We explore this concept in terms of social networks and illustrate some important results taken from the mathematical literature. In addition we show how the concept can be extended to ecological and perfect colorations, and discuss how the CATREGE algorithm can be extended to find the maximal exact regular coloration of a graph.

A multilevel algorithm for force-directed graph-drawing

We describe a heuristic method for drawing graphs which uses a multilevel framework combined with a force-directed placement algorithm. The multilevel technique matches and coalesces pairs of adjacent vertices to define a new graph and is repeated recursively to create a hierarchy of increasingly coarse graphs, G0, G1, …, GL. The coarsest graph, GL, is then given an initial layout and the layout is refined and extended to all the graphs starting with the coarsest and ending with the original. At each successive change of level, l, the initial layout for Gl is taken from its coarser and smaller child graph, Gl+1, and refined using force-directed placement. In this way the multilevel framework both accelerates and appears to give a more global quality to the drawing. The algorithm can compute both 2 & 3 dimensional layouts and we demonstrate it on examples ranging in size from 10 to 225,000 vertices. It is also very fast and can compute a 2D layout of a sparse graph in around 12 seconds for a 10,000 vertex graph to around 5-7 minutes for the largest graphs. This is an order of magnitude faster than recent implementations of force-directed placement algorithms.

Ghost and self: Jung's paradigm shift and a response to Zinkin

Zinkin's lucid challenge to Jung makes perfect sense. Indeed, it is the implications of this `making sense' that this paper addresses. For Zinkin's characterization of the `self' takes it as a `concept' requiring coherence; a variety of abstract non-contextual knowledge that itself has a mythical heritage. Moreover, Zinkin's refinement of Jung seeks to make his work fit for the scientific paradigm of modernity. In turn, modernity's paradigm owes much to Newton's notion of knowledge via reductionism. Here knowledge or investigation is divided up into the smallest possible units with the aim of eventually putting it all together into `one' picture of scientific truth. Unfortunately, `reductionism' does not do justice to the resonant possibilities of Jung's writing. These look forward to a new scientific paradigm of the twenty-first century, of the interactive `field', emergence and complexity theory. The paper works paradoxically by discovering Zinkin's `intersubjective self' after all, in two undervalued narratives by Jung, his doctoral thesis and a short late ghost story. However, in the ambivalences and radical fictional experimentation of these fascinating texts can be discerned an-Other self, one both created and found. [From the Publisher]

Adaptive ordering for imperfect successive decision feedback multiuser detection

A method for selecting the order in which the users are detected in communication systems employing adaptive successive decision feedback multiuser detection is proposed. Systems employing channel coding without the assumption of perfect decision feedback are analyzed. The method is based on the mean squared error (MSE) measurements during a training period for each user. The analysis' shows that the method delivers BER performance improvement relative to other previously proposed ordering methods