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.

A scalable strategy for the parallelization of multiphysics unstructured mesh-iterative codes on distributed-memory systems

Realizing scalable performance on high performance computing systems is not straightforward for single-phenomenon codes (such as computational fluid dynamics [CFD]). This task is magnified considerably when the target software involves the interactions of a range of phenomena that have distinctive solution procedures involving different discretization methods. The problems of addressing the key issues of retaining data integrity and the ordering of the calculation procedures are significant. A strategy for parallelizing this multiphysics family of codes is described for software exploiting finite-volume discretization methods on unstructured meshes using iterative solution procedures. A mesh partitioning-based SPMD approach is used. However, since different variables use distinct discretization schemes, this means that distinct partitions are required; techniques for addressing this issue are described using the mesh-partitioning tool, JOSTLE. In this contribution, the strategy is tested for a variety of test cases under a wide range of conditions (e.g., problem size, number of processors, asynchronous / synchronous communications, etc.) using a variety of strategies for mapping the mesh partition onto the processor topology.

CAPLib - A 'thin layer' message passing library to support computational mechanics codes on distributed memory parallel systems

The Computer Aided Parallelisation Tools (CAPTools) [Ierotheou, C, Johnson SP, Cross M, Leggett PF, Computer aided parallelisation tools (CAPTools)-conceptual overview and performance on the parallelisation of structured mesh codes, Parallel Computing, 1996;22:163±195] is a set of interactive tools aimed to provide automatic parallelisation of serial FORTRAN Computational Mechanics (CM) programs. CAPTools analyses the user's serial code and then through stages of array partitioning, mask and communication calculation, generates parallel SPMD (Single Program Multiple Data) messages passing FORTRAN. The parallel code generated by CAPTools contains calls to a collection of routines that form the CAPTools communications Library (CAPLib). The library provides a portable layer and user friendly abstraction over the underlying parallel environment. CAPLib contains optimised message passing routines for data exchange between parallel processes and other utility routines for parallel execution control, initialisation and debugging. By compiling and linking with different implementations of the library, the user is able to run on many different parallel environments. Even with today's parallel systems the concept of a single version of a parallel application code is more of an aspiration than a reality. However for CM codes the data partitioning SPMD paradigm requires a relatively small set of message-passing communication calls. This set can be implemented as an intermediate `thin layer' library of message-passing calls that enables the parallel code (especially that generated automatically by a parallelisation tool such as CAPTools) to be as generic as possible. CAPLib is just such a `thin layer' message passing library that supports parallel CM codes, by mapping generic calls onto machine specific libraries (such as CRAY SHMEM) and portable general purpose libraries (such as PVM an MPI). This paper describe CAPLib together with its three perceived advantages over other routes: - as a high level abstraction, it is both easy to understand (especially when generated automatically by tools) and to implement by hand, for the CM community (who are not generally parallel computing specialists); - the one parallel version of the application code is truly generic and portable; - the parallel application can readily utilise whatever message passing libraries on a given machine yield optimum performance.

SMARTFIRE: an integrated computational fluid dynamics code and expert system for fire field modelling

SMARTFIRE, an open architecture integrated CFD code and knowledge based system attempts to make fire field modeling accessible to non-experts in Computational Fluid Dynamics (CFD) such as fire fighters, architects and fire safety engineers. This is achieved by embedding expert knowledge into CFD software. This enables the 'black-art' associated with the CFD analysis such as selection of solvers, relaxation parameters, convergence criteria, time steps, grid and boundary condition specification to be guided by expert advice from the software. The user is however given the option of overriding these decisions, thus retaining ultimate control. SMARTFIRE also makes use of recent developments in CFD technology such as unstructured meshes and group solvers in order to make the CFD analysis more efficient. This paper describes the incorporation within SMARTFIRE of the expert fire modeling knowledge required for automatic problem setup and mesh generation as well as the concept and use of group solvers for automatic and manual dynamic control of the CFD code.

The Shared Memory Parallelisation of an Ocean Modelling Code using an Interactive Parallelisation Toolkit

This paper briefly describes an interactive parallelisation toolkit that can be used to generate parallel code suitable for either a distributed memory system (using message passing) or a shared memory system (using OpenMP). This study focuses on how the toolkit is used to parallelise a complex heterogeneous ocean modelling code within a few hours for use on a shared memory parallel system. The generated parallel code is essentially the serial code with OpenMP directives added to express the parallelism. The results show that substantial gains in performance can be achieved over the single thread version with very little effort.

Performance evaluation of a distributed algorithm for an inverse heat conduction problem

Numerical solutions of realistic 2-D and 3-D inverse problems may require a very large amount of computation. A two-level concept on parallelism is often used to solve such problems. The primary level uses the problem partitioning concept which is a decomposition based on the mathematical/physical problem. The secondary level utilizes the widely used data partitioning concept. A theoretical performance model is built based on the two-level parallelism. The observed performance results obtained from a network of general purpose Sun Sparc stations are compared with the theoretical values. Restrictions of the theoretical model are also discussed.

The Daedalus system: A tool for the Italian investigating magistrate

Daedalus is a computer tool, developed by an Italian magistrate - Carmelo Asaro - and integrated in his own daily routine as an investigating magistrate conducting inquiries, then as a prosecutor if and when the case investigated goes to court. This tool has recently been adopted by magistrates in judiciary offices throughout Italy, spawning moreover other related projects. First, this paper describes a sample session with daedalus. Next, an overview of an array of judicial tools leads to positioning daedalus in the context of the spectrum.

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.

Using an interactive parallelisation toolkit to parallelise an ocean modelling code

This paper describes an interactive parallelisation toolkit that can be used to generate parallel code suitable for either a distributed memory system (using message passing) or a shared memory system (using OpenMP). This study focuses on how the toolkit is used to parallelise a complex heterogeneous ocean modelling code within a few hours for use on a shared memory parallel system. The generated parallel code is essentially the serial code with OpenMP directives added to express the parallelism. The results show that substantial gains in performance can be achieved over the single thread version with very little effort.

Models and tools for an integrated European environmental management and decision support system, (IEEMDSS)

This work proceeds from the assumption that a European environmental information and communication system (EEICS) is already established. In the context of primary users (land-use planners, conservationists, and environmental researchers) we ask what use may be made of the EEICS for building models and tools which is of use in building decision support systems for the land-use planner. The complex task facing the next generation of environmental and forest modellers is described, and a range of relevant modelling approaches are reviewed. These include visualization and GIS; statistical tabulation and database SQL, MDA and OLAP methods. The major problem of noncomparability of the definitions and measures of forest area and timber volume is introduced and the possibility of a model-based solution is considered. The possibility of using an ambitious and challenging biogeochemical modelling approach to understanding and managing European forests sustainably is discussed. It is emphasised that all modern methodological disciplines must be brought to bear, and a heuristic hybrid modelling approach should be used so as to ensure that the benefits of practical empirical modelling approaches are utilised in addition to the scientifically well-founded and holistic ecosystem and environmental modelling. The data and information system required is likely to end up as a grid-based-framework because of the heavy use of computationally intensive model-based facilities.

Simulating the Swissair flight 111 in-flight fire using the CFD fire simulation software SMARTFIRE

At 8.18pm on 2 September 1998, Swissair Flight 111 (SR 111), took off from New York’s JFK airport bound for Geneva, Switzerland. Tragically, the MD-11 aircraft never arrived. According to the crash investigation report, published on 27 March 2003, electrical arcing in the ceiling void cabling was the most likely cause of the fire that brought down the aircraft. No one on board was aware of the disaster unfolding in the ceiling of the aircraft and, when a strange odour entered the cockpit, the pilots thought it was a problem with the air-conditioning system. Twenty minutes later, Swissair Flight 111 plunged into the Atlantic Ocean five nautical miles southwest of Peggy’s Cove, Nova Scotia, with the loss of all 229 lives on board. In this paper, the Computational Fluid Dynamics (CFD) analysis of the in-flight fire that brought down SR 111 is described. Reconstruction of the wreckage disclosed that the fire pattern was extensive and complex in nature. The fire damage created significant challenges to identify the origin of the fire and to appropriately explain the heat damage observed. The SMARTFIRE CFD software was used to predict the “possible” behaviour of airflow as well as the spread of fire and smoke within SR 111. The main aims of the CFD analysis were to develop a better understanding of the possible effects, or lack thereof, of numerous variables relating to the in-flight fire. Possible fire and smoke spread scenarios were studied to see what the associated outcomes would be. This assisted investigators at Transportation Safety Board (TSB) of Canada, Fire & Explosion Group in assessing fire dynamics for cause and origin determination.

Issues in delivering multimedia content to mobile devices

The scalability of a computer system is its response to growth. It is also depended on its hardware, its operating system and the applications it is running. Most distributed systems technology today still depends on bus-based shared memory which do not scale well, and systems based on the grid or hypercube scheme requires significantly less connections than a full inter-connection that would exhibit a quadratic growth rate. The rapid convergence of mobile communication, digital broadcasting and network infrastructures calls for rich multimedia content that is adaptive and responsive to the needs of individuals, businesses and the public organisations. This paper will discuss the emergence of mobile Multimedia systems and provides an overview of the issues regarding design and delivery of multimedia content to mobile devices.

Applying CBR to measure similarity of software design structures

This paper examines different ways for measuring similarity between software design models for the purpose of software reuse. Current approaches to this problem are discussed and a set of suitable similarity metrics are proposed and evaluated. Work on the optimisation of weights to increase the competence of a CBR system is presented. A graph matching algorithm and associated metrics capturing the structural similarity between UML class diagrams is presented and demonstrated through an example case.

Using an interactive software environment for the parallelization of real-world scientific applications

The parallelization of real-world compute intensive Fortran application codes is generally not a trivial task. If the time to complete the parallelization is to be significantly reduced then an environment is needed that will assist the programmer in the various tasks of code parallelization. In this paper the authors present a code parallelization environment where a number of tools that address the main tasks such as code parallelization, debugging and optimization are available. The ParaWise and CAPO parallelization tools are discussed which enable the near automatic parallelization of real-world scientific application codes for shared and distributed memory-based parallel systems. As user involvement in the parallelization process can introduce errors, a relative debugging tool (P2d2) is also available and can be used to perform nearly automatic relative debugging of a program that has been parallelized using the tools. A high quality interprocedural dependence analysis as well as user-tool interaction are also highlighted and are vital to the generation of efficient parallel code and in the optimization of the backtracking and speculation process used in relative debugging. Results of benchmark and real-world application codes parallelized are presented and show the benefits of using the environment