High-performance parallel systems for data-intensive computing

Thesis (Ph.D.)--University of Washington, 2016-08

Dynamic mesh partitioning and load-balancing for parallel computational mechanics codes

In this Chapter we discuss the load-balancing issues arising in parallel mesh based computational mechanics codes for which the processor loading changes during the run. We briefly touch on geometric repartitioning ideas and then focus on different ways of using a graph both to solve the load-balancing problem and the optimisation problem, both locally and globally. We also briefly discuss whether repartitioning is always valid. Sample illustrative results are presented and we conclude that repartitioning is an attractive option if the load changes are not too dramatic and that there is a certain trade-off between partition quality and volume of data that the underlying application needs to migrate.

Export control implementation issues with respect to high-performance computers : hearing before Committee on Governmental Affairs, United States Senate, One Hundred Sixth Congress, second session, May 26, 2000.

Shipping list no.: 2000-0347-P.

High performance computer export controls : hearing before the Committee on Governmental Affairs, United States Senate, One Hundred Seventh Congress, first session, March 15, 2001.

Shipping list no.: 2001-0258-P.

A synthesizable hardware evolutionary algorithm design for unmanned aerial system real-time path planning

The main objective of this paper is to detail the development of a feasible hardware design based on Evolutionary Algorithms (EAs) to determine flight path planning for Unmanned Aerial Vehicles (UAVs) navigating terrain with obstacle boundaries. The design architecture includes the hardware implementation of Light Detection And Ranging (LiDAR) terrain and EA population memories within the hardware, as well as the EA search and evaluation algorithms used in the optimizing stage of path planning. A synthesisable Very-high-speed integrated circuit Hardware Description Language (VHDL) implementation of the design was developed, for realisation on a Field Programmable Gate Array (FPGA) platform. Simulation results show significant speedup compared with an equivalent software implementation written in C++, suggesting that the present approach is well suited for UAV real-time path planning applications.

Parallel mesh partitioning on distributed memory systems

The problem of deriving parallel mesh partitioning algorithms for mapping unstructured meshes to parallel computers is discussed in this chapter. In itself this raises a paradox - we seek to find a high quality partition of the mesh, but to compute it in parallel we require a partition of the mesh. In fact, we overcome this difficulty by deriving an optimisation strategy which can find a high quality partition even if the quality of the initial partition is very poor and then use a crude distribution scheme for the initial partition. The basis of this strategy is to use a multilevel approach combined with local refinement algorithms. Three such refinement algorithms are outlined and some example results presented which show that they can produce very high global quality partitions, very rapidly. The results are also compared with a similar multilevel serial partitioner and shown to be almost identical in quality. Finally we consider the impact of the initial partition on the results and demonstrate that the final partition quality is, modulo a certain amount of noise, independent of the initial partition.

Systolic building block for high performance recursive filtering

A novel bit level systolic array is presented that can be used as a building block in the construction of recursive digital filters. The circuit accepts bit-parallel input data, is pipelined at the bit level, and exhibits a very high throughput rate. The most important feature of the circuit is that it allows recursive operations to be implemented directly without incurring the large m cycle latency (where m is approximately the word length) normally associated with such systems. The use of this circuit in the construction of both first- and second-order IIR (infinite-impulse-response) filters is described.

Simplifying gene expression microarray comparative analysis.

Gene Expression Comparative Analysis allows bioinformatics researchers to discover the conserved or specific functional regulation of genes. This is achieved through comparisons between quantitative gene expression measurements obtained in different species on different platforms to address a particular biological system. Comparisons are made more difficult due to the need to map orthologous genes between species, pre-processing of data (normalization) and post-analysis (statistical and correlation analysis). In this paper we introduce a web-based software package called EXP-PAC which provides on line interfaces for database construction and query of data, and makes use of a high performance computing platform of computer clusters to run gene sequence mapping and normalization methods in parallel. Thus, EXP-PAC facilitates the integration of gene expression data for comparative analysis and the online sharing, retrieval and visualization of complex multi-specific and multi-platform gene expression results.

Parallel implementation of stochastic simulation for large scale cellular processes

Experimental and theoretical studies have shown the importance of stochastic processes in genetic regulatory networks and cellular processes. Cellular networks and genetic circuits often involve small numbers of key proteins such as transcriptional factors and signaling proteins. In recent years stochastic models have been used successfully for studying noise in biological pathways, and stochastic modelling of biological systems has become a very important research field in computational biology. One of the challenge problems in this field is the reduction of the huge computing time in stochastic simulations. Based on the system of the mitogen-activated protein kinase cascade that is activated by epidermal growth factor, this work give a parallel implementation by using OpenMP and parallelism across the simulation. Special attention is paid to the independence of the generated random numbers in parallel computing, that is a key criterion for the success of stochastic simulations. Numerical results indicate that parallel computers can be used as an efficient tool for simulating the dynamics of large-scale genetic regulatory networks and cellular processes

Multi-scale hybrid numerical simulation of the growth of high-aspect-ratio nanostructures

Recently, a variety high-aspect-ratio nanostructures have been grown and profiled for various applications ranging from field emission transistors to gene/drug delivery devices. However, fabricating and processing arrays of these structures and determining how changing certain physical parameters affects the final outcome is quite challenging. We have developed several modules that can be used to simulate the processes of various physical vapour deposition systems from precursor interaction in the gas phase to gas-surface interactions and surface processes. In this paper, multi-scale hybrid numerical simulations are used to study how low-temperature non-equilibrium plasmas can be employed in the processing of high-aspect-ratio structures such that the resulting nanostructures have properties suitable for their eventual device application. We show that whilst using plasma techniques is beneficial in many nanofabrication processes, it is especially useful in making dense arrays of high-aspect-ratio nanostructures.

Computational modelling of multi-physics and multi-scale processes in parallel

This paper provides an overview of the developing needs for simulation software technologies for the computational modelling of problems that involve combinations of interactions amongst varying physical phenomena over a variety of time and space scales. Computational modelling of such problems requires software tech1nologies that enable the mathematical description of the interacting physical phenomena together with the solution of the resulting suites of equations in a numerically consistent and compatible manner. This functionality requires the structuring of simulation modules for specific physical phenomena so that the coupling can be effectively represented. These multi-physics and multi-scale computations are very compute intensive and the simulation software must operate effectively in parallel if it is to be used in this context. An approach to these classes of multi-disciplinary simulation in parallel is described, with some key examples of application to2 challenging engineering problems.

Multi-physics modelling of materials processes on high performance parallel computers

Abstract not available

Visualisation modelling in 3D of an alluvial aquifer system at a valley-wide scale to understand groundwater status : Lockyer, QLD

Visualisation provides a method to efficiently convey and understand the complex nature and processes of groundwater systems. This technique has been applied to the Lockyer Valley to aid in comprehending the current condition of the system. The Lockyer Valley in southeast Queensland hosts intensive irrigated agriculture sourcing groundwater from alluvial aquifers. The valley is around 3000 km2 in area and the alluvial deposits are typically 1-3 km wide and to 20-35 m deep in the main channels, reducing in size in subcatchments. The configuration of the alluvium is of a series of elongate “fingers”. In this roughly circular valley recharge to the alluvial aquifers is largely from seasonal storm events, on the surrounding ranges. The ranges are overlain by basaltic aquifers of Tertiary age, which overall are quite transmissive. Both runoff from these ranges and infiltration into the basalts provided ephemeral flow to the streams of the valley. Throughout the valley there are over 5,000 bores extracting alluvial groundwater, plus lesser numbers extracting from underlying sandstone bedrock. Although there are approximately 2500 monitoring bores, the only regularly monitored area is the formally declared management zone in the lower one third. This zone has a calibrated Modflow model (Durick and Bleakly, 2000); a broader valley Modflow model was developed in 2002 (KBR), but did not have extensive extraction data for detailed calibration. Another Modflow model focused on a central area river confluence (Wilson, 2005) with some local production data and pumping test results. A recent subcatchment simulation model incorporates a network of bores with short-period automated hydrographic measurements (Dvoracek and Cox, 2008). The above simulation models were all based on conceptual hydrogeological models of differing scale and detail.