The development of a data-driven application benchmarking approach to performance modelling

Performance modelling is a useful tool in the lifeycle of high performance scientific software, such as weather and climate models, especially as a means of ensuring efficient use of available computing resources. In particular, sufficiently accurate performance prediction could reduce the effort and experimental computer time required when porting and optimising a climate model to a new machine. In this paper, traditional techniques are used to predict the computation time of a simple shallow water model which is illustrative of the computation (and communication) involved in climate models. These models are compared with real execution data gathered on AMD Opteron-based systems, including several phases of the U.K. academic community HPC resource, HECToR. Some success is had in relating source code to achieved performance for the K10 series of Opterons, but the method is found to be inadequate for the next-generation Interlagos processor. The experience leads to the investigation of a data-driven application benchmarking approach to performance modelling. Results for an early version of the approach are presented using the shallow model as an example.

Achieving accurate FTIR measurements on high performance bandpass filters

The sources of ordinate error in FTIR spectrometers are reviewed with reference to measuring small out-of-band features in the spectra of bandpass filters. Procedures for identifying instrumental artefacts are described. It is shown that features well below 0.01%T can be measured reliably.

PRISM and ENES: a European approach to Earth system modelling

Europe's widely distributed climate modelling expertise, now organized in the European Network for Earth System Modelling (ENES), is both a strength and a challenge. Recognizing this, the European Union's Program for Integrated Earth System Modelling (PRISM) infrastructure project aims at designing a flexible and friendly user environment to assemble, run and post-process Earth System models. PRISM was started in December 2001 with a duration of three years. This paper presents the major stages of PRISM, including: (1) the definition and promotion of scientific and technical standards to increase component modularity; (2) the development of an end-to-end software environment (graphical user interface, coupling and I/O system, diagnostics, visualization) to launch, monitor and analyse complex Earth system models built around state-of-art community component models (atmosphere, ocean, atmospheric chemistry, ocean bio-chemistry, sea-ice, land-surface); and (3) testing and quality standards to ensure high-performance computing performance on a variety of platforms. PRISM is emerging as a core strategic software infrastructure for building the European research area in Earth system sciences. Copyright (c) 2005 John Wiley & Sons, Ltd.

High performance blocking filters for the region 1µm to 20µm

A set of filters based on the sequence of semiconductor edges is described which offers continuity of short-wave infrared blocking. The rejection throughout the stop region is greater than 103 for each filter and the transmission better than 70% through one octave with a square cutoff. The cutoff points are located at intervals of about two-thirds of an octave. Filters at 2.6 ,µm, 5.5 µm, and 12 µm which use a low-passing multilayer in combination with a semiconductor absorption edge are described in detail. The design of multilayers for optimum performance is discussed by analogy with the synthesis of electric circuit filters.

High performance subgraph mining in molecular compounds

Structured data represented in the form of graphs arises in several fields of the science and the growing amount of available data makes distributed graph mining techniques particularly relevant. In this paper, we present a distributed approach to the frequent subgraph mining problem to discover interesting patterns in molecular compounds. The problem is characterized by a highly irregular search tree, whereby no reliable workload prediction is available. We describe the three main aspects of the proposed distributed algorithm, namely a dynamic partitioning of the search space, a distribution process based on a peer-to-peer communication framework, and a novel receiver-initiated, load balancing algorithm. The effectiveness of the distributed method has been evaluated on the well-known National Cancer Institute’s HIV-screening dataset, where the approach attains close-to linear speedup in a network of workstations.

The potential for high performance design adoption in retail property portfolios

Growing legislative pressures and increasing stakeholder awareness of environmental issues are pushing the property market to consider high-performance, low-impact retail buildings. The office sector is relatively advanced in its apparent appreciation of such buildings; however, the retail sector is slow to recognize these benefits. In exploring the business case for high-performance design adoption in the retail sector, this paper examines the overlaps between office and retail sector benefits and considers the potential benefits peculiar to retailers. Barriers to high-performance design adoption are then addressed through case research, interviews with key representatives from the retail property market and a questionnaire survey of FTSE listed retail company property departments. The paper concludes that information gaps are a significant hindrance to high-performance property development and that they can be reduced, to some extent, by the forthcoming introduction of the BREEAM Retail environmental assessment tool. Copyright © 2003 John Wiley & Sons, Ltd and ERP Environment.

A novel approach to processing high-performance polymers that exploits entropically driven ring-opening polymerization

Blends of the poly(ether sulfone) derived from 4,4'-biphenol and 4,4'-dichlorodiphenylsulfone (Radel-R(TM)) with its homologous macrocyclic oligomers show greatly lowered melt viscosities relative to that of the parent polymer, potentially enabling more facile production and fabrication of fiber-reinforced composite materials. The macrocycles can then undergo entropically driven ring-opening polymerization in situ. The required blends can be obtained easily in one step, by carrying out polycondensations at concentrations lower than those usually used for polymer synthesis.

Ring-chain interconversion in high-performance polymer systems. 3. Cyclodepolymerization of poly(m-phenylene isophthalamide) (Nomex) and entropically driven ring-opening polymerization of the macrocyclic oligomers so produced

A homologous series of macrocyclic oligoamides has been prepared in high yield by reaction of isophthaloyl chloride with m-phenylenediamine under pseudo-high-dilution conditions. The products were characterized by infrared and H-1 NMR spectroscopies, matrix assisted laser desorption-ionization time-of-flight mass spectrometry, and gel permeation chromatography (GPC). A series of linear oligomers was prepared for comparison. The macrocycles ranged in size from the cyclic trimer up to at least the cyclic nonamer (90 ring atoms). The same homologous series of macrocyclic oligomers was prepared in high yield by the cyclodepolymerization of poly(m-phenylene isophthalamide) (Nomex). Cyclodepolymerization was best achieved by treating a 1% w/v solution of the polymer in dimethyl sulfoxide containing calcium chloride or lithium chloride with 3-4 mol % of sodium hydride or the sodium salt of benzanilide at 150 degreesC for 70 h. Treatment of a concentrated solution of the macrocyclic oligomers (25% w/v) with 4 mol % of sodium hydride or the sodium salt of benzanilide in a solution of lithium chloride in dimethyl sulfoxide at 170 degreesC for 6 h resulted in efficient entropically driven ring-opening polymerizations to give poly(m-phenylene isophthalamide), characterized by infrared and H-1 NMR spectroscopies and by GPC. The molecular weights obtained were comparable with those of the commercial polymer.

Microfabrication of high-performance aromatic polymers as nanotubes or fibrils by in situ ring-opening polymerisation of macrocyclic precursors

Melt-phase nucleophilic ring-opening polymerisation of macrocyclic aromatic ethers and thioethers at high temperatures within the cylindrical pores of an anodic-alumina membrane, followed by dissolution of the template, enables replication of the membrane's internal pore structure and so affords high-performance aromatic polymers with well-defined fibrillar or tubular morphologies.

Analysis of 'APN' naphthenic acids by high temperature gas chromatoagraphy and high performance liquid chromatography

Examination by high temperature GC (HTGC) of the methyl esters of the so-called 'ARN' naphthenic acids from crude oils of North Sea UK, Norwegian Sea and West African oilfields revealed the distributions of resolved 4-8 ring C-80 tetra acids and trace amounts of other acids. Whilst all three oils contained apparently the same the proportions of each differed, possibly reflecting the growth tempe acids, ratures of the archaebacteria from which the acids are assumed to have originated. The structures of the 4, 5, 7 and 8 ring acids are tentatively assigned by comparison with the known 6 ring acid and related natural products and an HPLC method for the isolation of the individual acids is described. ESI-MS of individual acids isolated by preparative HPLC established the elution order of the 4-8 ring acids on the HPLC and HTGC systems and revealed the presence of previously unreported acids tentatively identified as C-81 and C-82 7 and 8 ring analogues.

An optimized reverse-phase high performance liquid chromatographic method for evaluating percutaneous absorption of glucosamine hydrochloride

A relatively simple, selective, precise and accurate high performance liquid chromatography (HPLC) method based on a reaction of phenylisothiocyanate (PITC) with glucosamine (GL) in alkaline media was developed and validated to determine glucosamine hydrochloride permeating through human skin in vitro. It is usually problematic to develop an accurate assay for chemicals traversing skin because the excellent barrier properties of the tissue ensure that only low amounts of the material pass through the membrane and skin components may leach out of the tissue to interfere with the analysis. In addition, in the case of glucosamine hydrochloride, chemical instability adds further complexity to assay development. The assay, utilising the PITC-GL reaction was refined by optimizing the reaction temperature, reaction time and PITC concentration. The reaction produces a phenylthiocarbarnyl-glucosamine (PTC-GL) adduct which was separated on a reverse-phase (RP) column packed with 5 mu m ODS (C-18) Hypersil particles using a diode array detector (DAD) at 245 nm. The mobile phase was methanol-water-glacial acetic acid (10:89.96:0.04 v/v/v, pH 3.5) delivered to the column at 1 ml min(-1) and the column temperature was maintained at 30 degrees C Using a saturated aqueous solution of glucosamine hydrochloride, in vitro permeation studies were performed at 32 +/- 1 degrees C over 48 h using human epidermal membranes prepared by a heat separation method and mounted in Franz-type diffusion cells with a diffusional area 2.15 +/- 0.1 cm(2). The optimum derivatisation reaction conditions for reaction temperature, reaction time and PITC concentration were found to be 80 degrees C, 30 min and 1 % v/v, respectively. PTC-Gal and GL adducts eluted at 8.9 and 9.7 min, respectively. The detector response was found to be linear in the concentration range 0-1000 mu g ml(-1). The assay was robust with intra- and inter-day precisions (described as a percentage of relative standard deviation, %R.S.D.) < 12. Intra- and inter-day accuracy (as a percentage of the relative error, %RE) was <=-5.60 and <=-8.00, respectively. Using this assay, it was found that GL-HCI permeates through human skin with a flux 1.497 +/- 0.42 mu g cm(-2) h(-1), a permeability coefficient of 5.66 +/- 1.6 x 10(-6) cm h(-1) and with a lag time of 10.9 +/- 4.6 h. (c) 2005 Elsevier B.V. All rights reserved.

Parallel and distributed computing with Java

The Java language first came to public attention in 1995. Within a year, it was being speculated that Java may be a good language for parallel and distributed computing. Its core features, including being objected oriented and platform independence, as well as having built-in network support and threads, has encouraged this view. Today, Java is being used in almost every type of computer-based system, ranging from sensor networks to high performance computing platforms, and from enterprise applications through to complex research-based.simulations. In this paper the key features that make Java a good language for parallel and distributed computing are first discussed. Two Java-based middleware systems, namely MPJ Express, an MPI-like Java messaging system, and Tycho, a wide-area asynchronous messaging framework with an integrated virtual registry are then discussed. The paper concludes by highlighting the advantages of using Java as middleware to support distributed applications.