Accelerating an MPI Lattice Boltzmann code using OpenACC

Lattice Boltzmann method (LBM) is a promising approach to solving Computational Fluid Dynamics (CFD) problems, however, its nature of memory-boundness limits nearly all LBM algorithms' performance on modern computer architectures. This paper introduces novel sequential and parallel 3D memory-aware LBM algorithms to optimize its memory access performance. The introduced new algorithms combine the features of single-copy distribution, single sweep, swap algorithm, prism traversal, and merging two temporal time steps. We also design a parallel methodology to guarantee thread safety and reduce synchronizations in the parallel LBM algorithm. At last, we evaluate their performances on three high-end manycore systems and demonstrate that our new 3D memory-aware LBM algorithms outperform the state-of-the-art Palabos software (which realizes the Fuse Swap Prism LBM solver) by up to 89%.

Scientific collaboration is critical in high performance computational (HPC) environments. Computer security incidents, however, are a constant threat in today's interconnected computational environments. Many sites struggle with finding a balance between the needs of users and the threat from unauthorized usage or access. Understanding the constantly changing battlefield is critical for effective and optimal deployment of computer security resources.This tutorial will provide an underlying knowledge of the current state of computer security, its effects on HPC environments and mitigation strategies. Topics we will address include:1.HPC Security Fundamentals2.HPC Security Protection Fundamentals3.Intrusion Detection Using Bro4.The Changing Computer Protection Environment5.Cross-site Implications of Collaborative Environments6.Keeping One Step Ahead7.Risk Assessment, Mitigation and Compliance8.Components of a Good Protection ProgramWe will also show how described techniques are in use at SC06 to protect SCinet.

The efficient implementation of a diffraction tomography method for medical imaging is addressed within the framework of High Performance Computing (HPC) environment. A non-linear optimization method for the solution of the inverse scattering problem is implemented on a shared memory model computer. Linear speed-up and significant reduction in the total execution time is achieved when the program is executed in parallel, enabling the feasibility of the method for realistic medical imaging applications.

Purpose The present paper aims to address the description of a numerical optimization procedure, based on mesh morphing, and its application for the improvement of the aerodynamic performance of an industrial glider which suffers of a large separation occurring in the wing–fuselage junction region at high incidence angles. Design/methodology/approach Shape variations were applied to the baseline configuration through a mesh morphing technique founded on the mathematical framework of radial basis functions (RBF). The aerodynamic solutions were obtained coupling an RANS code with the mesh morphing tool RBF Morph™. Two shape modifiers were set up to generate a parametric numerical model. An optimization procedure, based on a design of experiment sampling, was set up implementing the fully automated workflow within a high performance computing (HPC) environment. The optimal candidates maximizing the aerodynamic efficiency were identified by means of a cubic RBF response surface approach. Findings The separation was significantly reduced, modifying the local geometry of fuselage and fairing and maintaining the wing aerofoil unchanged. A relevant aerodynamic efficiency improvement was finally gained. Practical implications The developed procedure proved to be a very powerful and efficient tool in facing aerodynamic design problems. However, it might be computationally very expensive if a large number of design variables are adopted and, in those cases, the method can be suitably used only within the HPC environment. Originality/value Such an optimization study is part of an explorative set of analyses that focused on better addressing the numerical strategies to be used in the development of the EU FP7 Project RBF4AERO.

Lattice Boltzmann method (LBM) is a new attractive computational approach for simulating isothermal multi-phase flows in computational fluid dynamics (CFD). It is based on the kinetic theory and easy to be parallelized. This study aims to analyze the performance of parallel LBM programming for the incompressible two-phase flows at high density and viscosity ratio. For this purpose, a liquid drop impact on a wetted wall with a pre-existing thin film of the same liquid is simulated by using the parallel LBM code. During the simulations, the domain decomposition, data communication and parallelization of the LBM code using the message passing interface (MPI) library have been investigated. The computational results show that the parallel LBM code exhibits a good high performance computing (HPC) on the parallel speed-up.

The lightweight nature, application portability, and deployment flexibility of containers is driving their widespread adoption in cloud solutions. Data analysis and deep learning (DL)/machine learning (ML) applications have especially benefited from containerization. As such data analysis is adopted in high performance computing (HPC), the need for container support in HPC has become paramount. However, containers face crucial performance and I/O challenges in HPC. One obstacle is that while there have been HPC containers, such solutions have not been thoroughly investigated, especially from the aspect of their impact on the crucial HPC I/O throughput. To this end, this paper provides a first-of-its-kind empirical analysis of state-of-the-art representative container solutions (Docker, Podman, Singularity, and Charliecloud) in HPC environments. We also explore how containers interact with an HPC parallel file system like Lustre. We present the design of an analysis framework that is deployed on all nodes in an HPC environment, and captures CPU, memory, network, and file I/O statistics from the nodes and the storage system. We are able to garner key insights from our analysis, e.g., Charliecloud outperforms other container solutions in terms of container start-up time, while Singularity and Charliecloud are equivalent in I/O throughput. But this comes at a cost, as Charliecloud invokes the most metadata and I/O operations on the underlying Lustre file system. By identifying such trade-offs and optimization opportunities, we can enhance HPC containers performance and the ML/DL applications that increasingly rely on them.

SummaryThe wide range of RNA-seq applications and their high-computational needs require the development of pipelines orchestrating the entire workflow and optimizing usage of available computational resources. We present aRNApipe, a project-oriented pipeline for processing of RNA-seq data in high-performance cluster environments. aRNApipe is highly modular and can be easily migrated to any high-performance computing (HPC) environment. The current applications included in aRNApipe combine the essential RNA-seq primary analyses, including quality control metrics, transcript alignment, count generation, transcript fusion identification, alternative splicing and sequence variant calling. aRNApipe is project-oriented and dynamic so users can easily update analyses to include or exclude samples or enable additional processing modules. Workflow parameters are easily set using a single configuration file that provides centralized tracking of all analytical processes. Finally, aRNApipe incorporates interactive web reports for sample tracking and a tool for managing the genome assemblies available to perform an analysis.Availability and documentationhttps://github.com/HudsonAlpha/aRNAPipe; DOI: 10.5281/zenodo.202950Supplementary informationSupplementary data are available at Bioinformatics online.

Abstract. Most Earth system models (ESMs) are running under different high-performance computing (HPC) environments. This has several advantages, from allowing different groups to work with the same tool in parallel to leveraging the burden of ensemble climate simulations, but it also offers alternative solutions in the case of shutdown (expected or not) of any of the environments. However, for obvious scientific reasons, it is critical to ensure that ESMs provide identical results under changes in computing environment. While strict bit-for-bit reproducibility is not always guaranteed with ESMs, it is desirable that results obtained under one computing environment are at least statistically indistinguishable from those obtained under another environment, which we term a “replicability” condition following the metrology nomenclature. Here, we develop a protocol to assess the replicability of the EC-Earth ESM. Using two versions of EC-Earth, we present one case of non-replicability and one case of replicability. The non-replicable case occurs with the older version of the model and likely finds its origin in the treatment of river runoff along Antarctic coasts. By contrast, the more recent version of the model provides replicable results. The methodology presented here has been adopted as a standard test by the EC-Earth consortium (27 institutions in Europe) to evaluate the replicability of any new model version across platforms, including for CMIP6 experiments. To a larger extent, it can be used to assess whether other ESMs can safely be ported from one HPC environment to another for studying climate-related questions. Our results and experience with this work suggest that the default assumption should be that ESMs are not replicable under changes in the HPC environment, until proven otherwise.

98P - Standardized and compliant bio-informatics pipeline for neoepitope analysis in oncology clinical trials

This chapter reports on the merging of geospatial data transformation, high-performance computing (HPC), and cyberinfrastructure (CI) domains for map projection transformation through performance profiling and tuning of pRasterBlaster, a parallel map projection transformation program. pRasterBlaster is built on the desktop version of mapIMG. Profiling was employed in an effort to identify and resolve computational bottlenecks that could prevent the program from scaling to thousands of processors for map projection on large raster datasets. Performance evaluation of a parallel program is critical to achieving projection transformation as factors such as the number of processors, overhead of communications, and input/output (I/O) all contribute to efficiency in an HPC environment. Flaws in the workload distribution algorithm, in this reported work, could hardly be observed when the number of processors was small. Without being exposed to large-scale supercomputers through software integration efforts, such flaws might remain unidentified. Overall, the two computational bottlenecks highlighted in this chapter, workload distribution and data-dependent load balancing, showed that in order to produce scalable code, profiling is an important process and scaling tests are necessary to identify bottlenecks that are otherwise difficult to discover.

At first thought, one may not realize the need for a special issue of Parallel and Distributed Computing Practices focused on Cluster Computing when our community has already invested years of research and spent a wealth of resources on traditional Big Iron supercomputers, for instance, the SGI/Cray Origin and IBM SP. Academia, industry, and national laboratories are still using, and for the foreseeable future, will continue to use, supercomputers to solve both grand-challenge scale and high-throughput applications. Then why the interest and new popularity of research into clusters for high-performance parallel and distributed applications? Clusters incorporate the tremendous successes that are attributable to our parallel and distributed processing community. Over the next decade, clusters will span the entire range of high-performance computing platforms. Instead of being restricted solely to supercomputers for achieving high-performance computing, the spectrum of available platforms will spread out to Beowulf-class clusters, to superclusters, and on to supercomputers. Moving along this continuum represents increasing computational resources, which in turn allows the capability for solving larger classes of applications or instances of problems. Compute-bound applications on workstations often may take advantage of the extra horsepower provided by symmetric multiprocessors. Applications with even larger computational requirements and a high-degree of concurrency may use Beowulf clusters, that is, clusters that are home-built from mass-market, commercial off-the-shelf workstations and networks, for instance, PC's interconnected by Fast Ethernet. A supercluster may contain commodity SMP nodes interconnected with scalable, low-latency, high-bandwidth networks, and is integrated by a system vendor rather than the computational scientist. Applications with greater complexity, for example, grand challenges from the sciences, demand scalable systems with high-performance, interconnects, perhaps coupled with terascale I/O subsystems, and thus, are appropriate for these superclusters. Finally, supercomputers, the largest of the clusters, may provide the ultimate platform for the highest-fidelity models and simulations and data-intensive applications. At each step up this ladder of cluster architectures, the system cost increases approximately by an order of magnitude, but having the high-performance capability to solve the problem of interest may require an entry point at any of these levels, the Beowulf, or supercluster, or even supercomputer platform. Cluster computing with Beowulf- or superclusters, thus, may be one or two orders of magnitude less expensive than a supercomputer, and provide a cost-effective solution and capability that was not available on a workstation. But this cost comparison measures just the basic system price. What about the applications? As recently as a few years ago, the lack of a canonical high-performance architectural paradigm meant that migrating to a new computer system, even from the same vendor, typically required redesigning applications to efficiently use the high-performance system. Thus, the application must be redesigned with parallel algorithms and libraries that are optimized for each high-performance system. This process of porting, redesigning, optimizing, debugging, and analyzing, the application may in fact be prohibitively expensive, and certainly frustrating when the process is completed just in time to greet the next generation system! That was then, this is now. With the successful standardization of system-level interfaces, many of these issues no longer hinder our expedited use of these new, high-performance clusters. The message-passing interface (MPI) is a clear example of a standard that has contributed significantly to the success of cluster computing. Supercomputer vendors working together with the high-performance computing community embarked on a process to standardize the user-level interface and behavior for passing messages between concurrent tasks. Now, using ordinary sequential compilers and vendor-supplied or freely-available implementations of this message-passing interface, applications can be implemented efficiently on one supercomputer and ported with little or no effort to other parallel machines. The message-passing paradigm, thus, provides a reasonably good canonical high-performance model and methodology for implementing high-performance applications. An algorithm that exhibits efficient execution and scaling on one machine then is likely to perform well on another with little or no additional programming effort. Using this cluster computing paradigm, algorithms typically are optimized by minimizing the number of remote memory accesses (reducing the amount of messages passed), and carefully organizing the data placement to increase memory locality. Thus, high-performance algorithms and applications that have been optimized for supercomputers, using the same cluster paradigm, may be appropriate for Beowulf and superclusters. Cluster computing, realizing the importance of a standard high-performance programming environment that applies equally well to both ends of the architectural spectrum, highly leverages these and other successes from parallel and distributed processing research and development. This is in an exciting, new era for high-performance computing where the canonical architectural paradigm stretches from home-built Beowulf clusters, to superclusters, and on to supercomputers. But cluster computing faces some challenges ahead of it. Continued research is required for designing 1) scalable, high-performance algorithms, applications, and methodologies, that can tolerate the communication overheads, and adapt to hybrid, SMP, and heterogeneous, nodes; 2) high-performance computer architectures, scalable interconnection networks, and I/O subsystems; and 3) scalable systems software, compilers, and monitoring tools. Cluster computing has come of age, and it is high time for the parallel and distributed processing community to lay the groundwork for our own legacy of success stories. David A. Bader University of New Mexico

Scientific workflows are models composed of activities, data and dependencies whose objective is to represent a computer simulation. Workflows are managed by Scientific Workflow Management System (SWfMS). Such workflows commonly demand for many computational resources once their executions may involve a number of different programs processing a huge volume of data. Thus, the use of High Performance Computing (HPC) environments allied to parallelization techniques provides the support for the execution of such experiments. Some resources provided by clouds can be used to build HPC environments. Although clouds offer advantages such as elasticity and availability, failures are a reality rather than a possibility. Thus, SWfMS must be fault-tolerant. There are several types of fault tolerance techniques used in SWfMS such as checkpoint-restart and replication, but which fault tolerance technique best fits with a specific workflow? This work aims at analyzing several fault tolerance techniques in SWfMSs and recommending the suitable one for the user’s workflow using machine learning techniques and provenance data, thus improving resiliency.

This chapter gives a comprehensive overview of the current status of accounting and billing for up-todate computing environments. Accounting is the key for the management of information system resources. At this stage of evolution of accounting systems it is adequate not to separate computing environments into High Performance Computing and Grid Computing environments for allowing a “holistic” view showing the different approaches and the state of the art for integrated accounting and billing in distributed computing environments. Requirements resulting from a public survey within all communities of the German Grid infrastructure, as well as from computing centres and resource providers of High Performance Computing resources like HLRN, and ZIVGrid, within the German e-Science framework, have been considered as well as requirements resulting from various information systems and the virtualisation of organisations and resources. Additionally, conceptual, technical, economical, and legal questions also had to be taken into consideration. After the requirements have been consolidated and implementations have been done over one year ago, now the overall results and conclusions are presented in the following sections showing a case study based on the GISIG framework and the Grid- GIS framework. The focus is on how an integrated architecture can be built and used in heterogeneous environments. A prototypical implementation is outlined that is able to manage and visualise relevant accounting and billing information based on suitable monitoring data in a virtual organisation (VO) specific way regarding basic business, economic, and security issues.

Modern High Performance Computing (HPC) applications, such as Earth science simulations, produce large amounts of data due to the surging of computing power, while big data applications have become more compute-intensive due to increasingly sophisticated analysis algorithms. The needs of both HPC and big data technologies for advanced HPC and big data applications create a demand for integrated system support. In this study, we introduce Scientific Data Processing (SciDP) to support both HPC and big data applications via integrated scientific data processing. SciDP can directly process scientific data stored on a Parallel File System (PFS), which is typically deployed in an HPC environment, in a big data programming environment running atop Hadoop Distributed File System (HDFS). SciDP seamlessly integrates PFS, HDFS, and the widely-used R data analysis system to support highly efficient processing of scientific data. It utilizes the merits of both PFS and HDFS for fast data transfer, overlaps computing with data accessing, and integrates R into the data transfer process. Experimental results show that SciDP accelerates analysis and visualization of a production NASA Center for Climate Simulation (NCCS) climate and weather application by 6x to 8x when compared to existing solutions.

Arc4nix: A cross-platform geospatial analytical library for cluster and cloud computing