High Performance Computing in Science and Engineering ’01

Probably the most impressive geological feature of active fault zones hosted in carbonate rocks is the presence of several hundreds of meters thick damage zones, often composed of in-situ shattered rocks (ISRs, i.e. rocks fragmented into clasts < 1 cm in size). Despite their abundance, it remains unknown how ISRs form (during the propagation of seismic ruptures?), and how their presence affects (1) the propagation of individual mainshock seismic ruptures, (2) the near field wave radiation and associated strong ground motions, and (3) the evolution in space and time of aftershock seismic sequences. In this contribution, we will present preliminary results of a three-year Ph.D. project aimed at addressing these issues through an integrated field geology and numerical modelling approach.We exploit existing and newly acquired field geology data on fault damage zone distributions in the Central Apennines (Italy), and perform dynamic rupture earthquake sequence simulations with SeisSol (https://seissol.org). The fully-dynamic individual earthquake simulations with SeisSol rely on the discontinuous Galerkin method, which allows treating complex 3D geological structures, nonlinear rheologies (including off-fault plastic yielding) and high-order accurate propagation of seismic waves (Käser et al., 2010). The earthquake modelling simulations integrate laboratory-derived frictional constitutive laws with simplified and realistic representations of fault zone geometry and surface topography. Currently, our study is focused on the 25 km long Campo Imperatore fault system in the Gran Sasso Massif area (Italian Central Apennines) where the damage zones are pronounced and well mapped (Demurtas et al., 2016; Fondriest et al., 2020).We aim at using the dynamic rupture earthquake modelling simulations to discuss the formation and distribution of ISRs with respect to (1) the maximum magnitude (Mw 7.0) of the earthquake associated with the studied fault, (2) fault geometry (length, presence of step overs, fault bends, etc.), (3) topographic effects (valleys, etc.), and (4) lithology (limestones, dolostones, etc.) of the wall rocks. This approach is expected to identify the physical, geological, and loading conditions controlling seismic rupture propagation and the development of fault damage zones. The physically based, fully dynamic 3D simulations will also provide estimates of earthquake source parameters (e.g., fracture energy and seismic moment release rate) and synthetic seismograms (strong ground motions), which will be compared with seismological and strong-motion data from earthquakes in the Central Apennines. References Demurtas, M., Fondriest, M., Balsamo, F., Clemenzi, L., Storti, F., Bistacchi, A., & Di Toro, G. (2016). Structure of a normal seismogenic fault zone in carbonates: The Vado di Corno Fault, Campo Imperatore, Central Apennines (Italy). Journal of Structural Geology, 90, 185–206. https://doi.org/10.1016/j.jsg.2016.08.004Fondriest, M., Balsamo, F., Bistacchi, A., Clemenzi, L., Demurtas, M., Storti, F., & Di Toro, G. (2020). Structural Complexity and Mechanics of a Shallow Crustal Seismogenic Source (Vado di Corno Fault Zone, Italy). Journal of Geophysical Research: Solid Earth, 125(9), e2019JB018926. https://doi.org/10.1029/2019JB018926Käser, M., Castro, C., Hermann, V., & Pelties, C. (2010). SeisSol – A Software for Seismic Wave Propagation Simulations. In High Performance Computing in Science and Engineering, Garching/Munich 2009 (pp. 281–292). Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-13872-0_24

High Performance Computing in Science and Engineering, Munich 2002

High Performance Computing in Science and Engineering '23

High Performance Computing in Science and Engineering ´16

High Performance Computing in Science and Engineering ’03

Combinatorial Scientific Computing explores the latest research on creating algorithms and software tools to solve key combinatorial problems on large-scale high-performance computing architectures. It includes contributions from international researchers who are pioneers in designing software and applications for high-performance computing systems. The book offers a state-of-the-art overview of the latest research, tool development, and applications. It focuses on load balancing and parallelization on high-performance computers, large-scale optimization, algorithmic differentiation of numerical simulation code, sparse matrix software tools, and combinatorial challenges and applications in large-scale social networks. The authors unify these seemingly disparate areas through a common set of abstractions and algorithms based on combinatorics, graphs, and hypergraphs. Combinatorial algorithms have long played a crucial enabling role in scientific and engineering computations and their importance continues to grow with the demands of new applications and advanced architectures. By addressing current challenges in the field, this volume sets the stage for the accelerated development and deployment of fundamental enabling technologies in high-performance scientific computing.

This paper reports on a four-year project that aims to raise the abstraction level through the use of model-driven engineering (MDE) techniques in the development of scientific applications relying on high-performance computing. The development and maintenance of high-performance scientific computing software is reputedly a complex task. This complexity results from the frequent evolutions of supercomputers and the tight coupling between software and hardware aspects. Moreover, current parallel programming approaches result in a mixing of concerns within the source code. Our approach relies on the use of MDE and consists in defining domain-specific modeling languages targeting various domain experts involved in the development of HPC applications, allowing each of them to handle their dedicated model in a both user-friendly and hardware-independent way. The different concerns are separated thanks to the use of several models as well as several modeling viewpoints on these models. Depending on the targeted execution platforms, these abstract models are translated into executable implementations by means of model transformations. To make all of these effective, we have developed a tool chain that is also presented in this paper. The approach is assessed through a multi-dimensional validation that focuses on its applicability, its expressiveness and its efficiency. To capitalize on the gained experience, we analyze some lessons learned during this project.

We describe our experiences with computational science and engineering (CS&E) education at the University of Colorado at Boulder in order to illustrate the difficulties associated with such a course of study. Our CS&E offerings began with a course on high-performance scientific computing (HPSC) developed under a CISE Educational Infrastructure grant from the National Science Foundation. The course was taught from 1991 through 1998 when various practical concerns ended its run. The course was revived in 2003 to meet the demands of several emerging programs in computational science. In this paper, we outline the rise, fall, and restoration of the HPSC course. We identify the technological developments that presently make such a course less labor intensive and more sustainable.

Fundamental transformations in the basic logic of computing are few and far between. Since the invention of digital computers in the early 1940s, the logic underlying computation has remained the same, even as computing hardware evolved from vacuum tubes to silicon transistors. With the advent of quantum computation, a fundamental transformation is near. Quantum computation is based on a different type of logic: rather than being in one of the two states of a classical bit, a quantum bit or qubit can be in a superposition of two states simultaneously. Operations and measurements on these qubits obey the constraints of quantum mechanics. It is now understood that quantum computers have great power in principle to go beyond classical computers, but that not every application is well suited to implementation on quantum computers. For reasons explained in more detail in the Introduction, scientific problems in chemical and materials sciences are uniquely suited to take advantage of quantum computing in the relatively near future. Indeed, quantum computing offers the best hope to solve many of the most important and difficult problems in this field. For example, quantum materials, such as superconductors and complex magnetic materials, show novel kinds of ordered phases that are natural from the point of view of quantum mechanics but difficult to access via computation on classical computers. Quantum sensors based on solid materials are already widely used but could be greatly improved with insight from quantum computations, as could materials for information technologies. Quantum chemical dynamics is another example of a problem that is intrinsically well suited to studies on quantum computers. Applications of quantum chemical dynamics include catalysis, artificial photosynthesis, and other industrially important processes. Quantum computers exist in the laboratory and are beginning to exceed 50 qubits, which is roughly the size beyond which their behavior cannot be predicted or emulated on present-day classical supercomputers. While a quantum computer of 50 qubits is almost certainly not powerful enough to tackle the major scientific challenges in chemical and materials sciences, some of these major challenges start to become accessible with a few hundred qubits if error rates can be kept small. This roundtable was convened to ask how emerging quantum computers can be applied to major scientific problems in chemical and materials sciences, in light of Basic Energy Sciences’s leading role in these fields and the Department of Energy’s leading role in high-performance scientific computation more generally. The main outcome of the roundtable was a consensus that there are scientific problems of great importance on which emerging quantum computers have the potential for disruptive impact, and where comparable progress is unlikely to occur by other means.

The Common Component Architecture (CCA) effort is the embodiment of a long-range program of research and development into the formulation, roles, and use of component technologies in high-performance scientific computing. CCA components can interoperate with other components in a variety of frameworks, including SCIRun2 from the University of Utah. The SCIRun2 framework is also developing the ability to connect components from a variety of different models through a mechanism called meta-components. The meta component model operates by providing a plugin architecture for component models. Abstract components are manipulated and managed by the SCIRun2 framework, while concrete component models perform the actual work and communicate with each other directly. We will leverage the SCIRun2 framework and the Kepler system to orchestrate components in the Fusion Simulation Project (FSP) and to provide a CCA-based interface with Kepler. The groundwork for this functionality is being performed with the Scientific Data Management center. The SDM center is developing CCA-compliant interfaces for expressing and executing workflows and create workflow components based on SCIRun and Ptolemy (Kepler) execution engines, including development of uniform interfaces for selecting, starting, and monitoring scientific workflows. Accomplishments include Introduction to CCA and Simulation Software Systems, Introduction into SCIRun2 and Bridging within SCIRun2, CCALoop: A scalable design for a distributed component framework, and Combining Workflow methodologies with Component Architectures.

GPGPU implementation of mixed spectral-finite difference computational code for the numerical integration of the three-dimensional time-dependent incompressible Navier–Stokes equations

Virtualization has become a common abstraction layer in modern data centers, enabling resource owners to manage complex infrastructure independently of their applications. Conjointly virtualization is becoming a driving technology for a manifold of industry grade IT services. The cloud concept includes the notion of a separation between resource owners and users, adding services such as hosted application frameworks and queuing. Utilizing the same infrastructure, clouds carry significant potential for use in high-performance scientific computing. The ability of clouds to provide for requests and releases of vast computing resource dynamically and close to the marginal cost of providing the services is unprecedented in the history of scientific and commercial computing. Distributed computing concepts that leverage federated resource access are popular within the grid community, but have not seen previously desired deployed levels so far. Also, many of the scientific datacenters have not adopted virtualization or cloud concepts yet. This workshop aims to bring together industrial providers with the scientific community in order to foster discussion, collaboration and mutual exchange of knowledge and experience. This year's workshop featured 9 papers on diverse topics in HPC virtualization. Papers of note include Kim et al. proposing group-based cloud memory deduplication along with Nanos et al. presenting results from a high-performance cluster interconnect prototype for VMs with a user-level RDMA protocol over standard 10Gbps Ethernet. The chairs would like to thank the Euro-Par organizers and the members of the program committee along with the speakers and attendees, whose interaction contributed to a stimulating environment. VHPC is planning to continue the successful co-location with Euro-Par in 2012.

The National Information Infrastructure of the future will be built upon core HPCC technologies such as parallel architectures, software, and high-speed networks. In addition to the key enabling technologies associated with high-performance scientific computing, such as image processing algorithms and linear algebra libraries, information processing applications will require development and integration of new, enabling HPCC technologies such as collaboration services, and multimedia databases. These new technologies will find application in information industries such as home entertainment and agile manufacturing.

The Service Oriented Architecture (SOA) is embraced in distributed and grid computing to produce high performance results for long time. SOA is likened by application programmers for its trademark characteristics of programmability, efficiency in heterogeneous conditions and fault-tolerance. It has worked well for high performance financial applications. but not for scientific applications which are too fine grained and communication intensive to be efficient on distributed environments. This paper argues that to make SOA model work well for those scientific applications, we need to reduce overhead costs associated with smaller task loads arising from finer granularity and increased communications in those applications. This paper proposes a data service to be used along with the existing compute services in SOA middlewares to enable inter-communication of finer tasks with out loosing SOA properties of programmability and efficiency under heterogeneity. This data service shall better enable high performance scientific computing of medium to fine grained scientific applications.

Online training has been growing in popularity, and offers many advantages for both trainers and learners. Assessing the usage and impact of online material can be difficult, especially if content is made available to anyone and is not part of a course requiring formal enrollment. The Cornell Virtual Workshop (CVW) first offered online training on topics in high-performance computing and computational science in 1994, and ten years ago we began logging usage. We are now performing our first in-depth analysis of those log data to identify patterns in usage, so that we can better understand how users access the material, which types of topics and materials result in the greatest impact, how topic usage changes over time, and what types of presentation format might be preferred. While the CVW is built around a cohesive, sequential narrative for each training topic, we find that many users access our content in a more targeted fashion, suggesting that we rethink how we package our material. We anticipate that ongoing analysis using data science and machine learning methods will enable us to produce more useful training materials, and provide the educational community with valuable information about patterns in online material usage.