MPJ express meets gadget: towards a java code for cosmological simulations

Gadget-2 is a massively parallel structure formation code for cosmological simulations. In this paper, we present a Java version of Gadget-2. We evaluated the performance of the Java version by running colliding galaxies simulation and found that it can achieve around 70% of C Gadget-2's performance.

A buffering layer to support derived types and proprietary networks for Java HPC

MPJ Express is our implementation of MPI-like bindings for Java. In this paper we discuss our intermediate buffering layer that makes use of the so-called direct byte buffers introduced in the Java New I/O package. The purpose of this layer is to support the implementation of derived datatypes. MPJ Express is the first Java messaging library that implements this feature using pure Java. In addition, this buffering layer allows efficient implementation of communication devices based on proprietary networks such as Myrinet. In this paper we evaluate the performance of our buffering layer and demonstrate the usefulness of direct byte buffers. Also, we evaluate the performance of MPJ Express against other messaging systems using Myrinet and show that our buffering layer has made it possible to avoid the overheads suffered by other Java systems such as mpiJava that relies on the Java Native Interface.

A buffering layer to support derived types and proprietary networks for Java HPC

MPJ Express is our implementation of MPI-like bindings for Java. In this paper we discuss our intermediate buffering layer that makes use of the so-called direct byte buffers introduced in the Java New I/O package. The purpose of this layer is to support the implementation of derived datatypes. MPJ Express is the first Java messaging library that implements this feature using pure Java. In addition, this buffering layer allows efficient implementation of communication devices based on proprietary networks such as Myrinet. In this paper we evaluate the performance of our buffering layer and demonstrate the usefulness of direct byte buffers. Also, we evaluate the performance of MPJ Express against other messaging systems using Myrinet and show that our buffering layer has made it possible to avoid the overheads suffered by other Java systems such as mpiJava that relies on the Java Native Interface.

Nested parallelism for multi-core HPC systems using Java

Since its introduction in 1993, the Message Passing Interface (MPI) has become a de facto standard for writing High Performance Computing (HPC) applications on clusters and Massively Parallel Processors (MPPs). The recent emergence of multi-core processor systems presents a new challenge for established parallel programming paradigms, including those based on MPI. This paper presents a new Java messaging system called MPJ Express. Using this system, we exploit multiple levels of parallelism - messaging and threading - to improve application performance on multi-core processors. We refer to our approach as nested parallelism. This MPI-like Java library can support nested parallelism by using Java or Java OpenMP (JOMP) threads within an MPJ Express process. Practicality of this approach is assessed by porting to Java a massively parallel structure formation code from Cosmology called Gadget-2. We introduce nested parallelism in the Java version of the simulation code and report good speed-ups. To the best of our knowledge it is the first time this kind of hybrid parallelism is demonstrated in a high performance Java application. (C) 2009 Elsevier Inc. All rights reserved.

A comparative study of Java and C performance in two large-scale parallel applications

In the 1990s the Message Passing Interface Forum defined MPI bindings for Fortran, C, and C++. With the success of MPI these relatively conservative languages have continued to dominate in the parallel computing community. There are compelling arguments in favour of more modern languages like Java. These include portability, better runtime error checking, modularity, and multi-threading. But these arguments have not converted many HPC programmers, perhaps due to the scarcity of full-scale scientific Java codes, and the lack of evidence for performance competitive with C or Fortran. This paper tries to redress this situation by porting two scientific applications to Java. Both of these applications are parallelized using our thread-safe Java messaging system—MPJ Express. The first application is the Gadget-2 code, which is a massively parallel structure formation code for cosmological simulations. The second application uses the finite-domain time-difference method for simulations in the area of computational electromagnetics. We evaluate and compare the performance of the Java and C versions of these two scientific applications, and demonstrate that the Java codes can achieve performance comparable with legacy applications written in conventional HPC languages. Copyright © 2009 John Wiley & Sons, Ltd.

Assessing the impact of nano- and micro-scale zerovalent iron particles on soil microbial activities: Particle reactivity interferes with assay conditions and interpretation of genuine microbial effects

The effects of nano-scale and micro-scale zerovalent iron (nZVI and mZVI) particles on general (dehydrogenase and hydrolase) and specific (ammonia oxidation potential, AOP) activities mediated by the microbial community in an uncontaminated soil were examined. nZVI (diameter 12.5 nm; 10 mg gÿ1 soil)apparently inhibited AOP and nZVI and mZVI apparently stimulated dehydrogenase activity but had minimal influence on hydrolase activity. Sterile experiments revealed that the apparent inhibition of AOP could not be interpreted as such due to the confounding action of the particles, whereas, the nZVIenhanced dehydrogenase activity could represent the genuine response of a stimulated microbial population or an artifact of ZVI reactivity. Overall, there was no evidence for negative effects of nZVI or mZVI on the processes studied. When examining the impact of redox active particles such as ZVI on microbial oxidation–reduction reactions, potential confounding effects of the test particles on assay conditions should be considered.

Multicore-enabling the MPJ express messaging library

With the transition to multicore processors almost complete, the parallel processing community is seeking efficient ways to port legacy message passing applications on shared memory and multicore processors. MPJ Express is our reference implementation of Message Passing Interface (MPI)-like bindings for the Java language. Starting with the current release, the MPJ Express software can be configured in two modes: the multicore and the cluster mode. In the multicore mode, parallel Java applications execute on shared memory or multicore processors. In the cluster mode, Java applications parallelized using MPJ Express can be executed on distributed memory platforms like compute clusters and clouds. The multicore device has been implemented using Java threads in order to satisfy two main design goals of portability and performance. We also discuss the challenges of integrating the multicore device in the MPJ Express software. This turned out to be a challenging task because the parallel application executes in a single JVM in the multicore mode. On the contrary in the cluster mode, the parallel user application executes in multiple JVMs. Due to these inherent architectural differences between the two modes, the MPJ Express runtime is modified to ensure correct semantics of the parallel program. Towards the end, we compare performance of MPJ Express (multicore mode) with other C and Java message passing libraries---including mpiJava, MPJ/Ibis, MPICH2, MPJ Express (cluster mode)---on shared memory and multicore processors. We found out that MPJ Express performs signicantly better in the multicore mode than in the cluster mode. Not only this but the MPJ Express software also performs better in comparison to other Java messaging libraries including mpiJava and MPJ/Ibis when used in the multicore mode on shared memory or multicore processors. We also demonstrate effectiveness of the MPJ Express multicore device in Gadget-2, which is a massively parallel astrophysics N-body siimulation code.

Reactive Halogens in the Marine Boundary Layer (RHaMBLe): the tropical North Atlantic experiments

The NERC UK SOLAS-funded Reactive Halogens in the Marine Boundary Layer (RHaMBLe) programme comprised three field experiments. This manuscript presents an overview of the measurements made within the two simultaneous remote experiments conducted in the tropical North Atlantic in May and June 2007. Measurements were made from two mobile and one ground-based platforms. The heavily instrumented cruise D319 on the RRS Discovery from Lisbon, Portugal to São Vicente, Cape Verde and back to Falmouth, UK was used to characterise the spatial distribution of boundary layer components likely to play a role in reactive halogen chemistry. Measurements onboard the ARSF Dornier aircraft were used to allow the observations to be interpreted in the context of their vertical distribution and to confirm the interpretation of atmospheric structure in the vicinity of the Cape Verde islands. Long-term ground-based measurements at the Cape Verde Atmospheric Observatory (CVAO) on São Vicente were supplemented by long-term measurements of reactive halogen species and characterisation of additional trace gas and aerosol species during the intensive experimental period. This paper presents a summary of the measurements made within the RHaMBLe remote experiments and discusses them in their meteorological and chemical context as determined from these three platforms and from additional meteorological analyses. Air always arrived at the CVAO from the North East with a range of air mass origins (European, Atlantic and North American continental). Trace gases were present at stable and fairly low concentrations with the exception of a slight increase in some anthropogenic components in air of North American origin, though NOx mixing ratios during this period remained below 20 pptv (note the non-IUPAC adoption in this manuscript of pptv and ppbv, equivalent to pmol mol−1 and nmol mol−1 to reflect common practice). Consistency with these air mass classifications is observed in the time series of soluble gas and aerosol composition measurements, with additional identification of periods of slightly elevated dust concentrations consistent with the trajectories passing over the African continent. The CVAO is shown to be broadly representative of the wider North Atlantic marine boundary layer; measurements of NO, O3 and black carbon from the ship are consistent with a clean Northern Hemisphere marine background. Aerosol composition measurements do not indicate elevated organic material associated with clean marine air. Closer to the African coast, black carbon and NO levels start to increase, indicating greater anthropogenic influence. Lower ozone in this region is possibly associated with the increased levels of measured halocarbons, associated with the nutrient rich waters of the Mauritanian upwelling. Bromide and chloride deficits in coarse mode aerosol at both the CVAO and on D319 and the continuous abundance of inorganic gaseous halogen species at CVAO indicate significant reactive cycling of halogens. Aircraft measurements of O3 and CO show that surface measurements are representative of the entire boundary layer in the vicinity both in diurnal variability and absolute levels. Above the inversion layer similar diurnal behaviour in O3 and CO is observed at lower mixing ratios in the air that had originated from south of Cape Verde, possibly from within the ITCZ. ECMWF calculations on two days indicate very different boundary layer depths and aircraft flights over the ship replicate this, giving confidence in the calculated boundary layer depth.

The north Atlantic marine oundary layer experiment (NAMBLEX): Overview of the campaign held at Mace Head, Ireland, in summer 2002

The North Atlantic Marine Boundary Layer Experiment (NAMBLEX), involving over 50 scientists from 12 institutions, took place at Mace Head, Ireland (53.32° N, 9.90° W), between 23 July and 4 September 2002. A wide range of state-of-the-art instrumentation enabled detailed measurements of the boundary layer structure and atmospheric composition in the gas and aerosol phase to be made, providing one of the most comprehensive in situ studies of the marine boundary layer to date. This overview paper describes the aims of the NAMBLEX project in the context of previous field campaigns in the Marine Boundary Layer (MBL), the overall layout of the site, a summary of the instrumentation deployed, the temporal coverage of the measurement data, and the numerical models used to interpret the field data. Measurements of some trace species were made for the first time during the campaign, which was characterised by predominantly clean air of marine origin, but more polluted air with higher levels of NOx originating from continental regions was also experienced. This paper provides a summary of the meteorological measurements and Planetary Boundary Layer (PBL) structure measurements, presents time series of some of the longer-lived trace species (O3, CO, H2, DMS, CH4, NMHC, NOx, NOy, PAN) and summarises measurements of other species that are described in more detail in other papers within this special issue, namely oxygenated VOCs, HCHO, peroxides, organo-halogenated species, a range of shorter lived halogen species (I2, OIO, IO, BrO), NO3 radicals, photolysis frequencies, the free radicals OH, HO2 and (HO2+Σ RO2), as well as a summary of the aerosol measurements. NAMBLEX was supported by measurements made in the vicinity of Mace Head using the NERC Dornier-228 aircraft. Using ECMWF wind-fields, calculations were made of the air-mass trajectories arriving at Mace Head during NAMBLEX, and were analysed together with both meteorological and trace-gas measurements. In this paper a chemical climatology for the duration of the campaign is presented to interpret the distribution of air-mass origins and emission sources, and to provide a convenient framework of air-mass classification that is used by other papers in this issue for the interpretation of observed variability in levels of trace gases and aerosols.

Twists and turns: a tale of two shikimate-pathway enzymes

We are studying two enzymes from the shikimate pathway, dehydroquinate synthase (DHQS) and 5-enolpyruvylshikimate-3-phosphate synthase (EPSPS). Both enzymes have been the subject of numerous studies to elucidate their reaction mechanisms. Crystal structures of DHQS and EPSPS in the presence and absence of substrates, cofactors and/or inhibitors are now available. These structures reveal movements of domains, rearrangements of loops and changes in side-chain positions necessary for the formation of a catalytically competent active site. The potential for using complementary small-angle X-ray scattering (SAXS) studies to confirm the presence of these structural differences in solution has also been explored. Comparative analysis of crystal structures, in the presence and absence of ligands, has revealed structural features critical for substrate-binding and catalysis. We have also analysed these structures by generating GRID energy maps to detect favourable binding sites. The combination of X-ray crystallography, SAXS and computational techniques provides an enhanced analysis of structural features important for the function of these complex enzymes.