A5102010274- Distributed and Parallel Computing Systems
- Scientific Computing and Data Management
- Advanced Data Storage Technologies
- Parallel Computing and Optimization Techniques
- Cloud Computing and Resource Management
- Scientific Research and Discoveries
- Electronic and Structural Properties of Oxides
- Reservoir Engineering and Simulation Methods
- Algorithms and Data Compression
- Nuclear reactor physics and engineering
- Machine Learning in Materials Science
- Genomics and Phylogenetic Studies
- Computational Physics and Python Applications
- Simulation Techniques and Applications
- Fusion materials and technologies
- Advanced Multi-Objective Optimization Algorithms
- Theoretical and Computational Physics
- Bioinformatics and Genomic Networks
- Advanced Condensed Matter Physics
- Bacteriophages and microbial interactions
- Distributed systems and fault tolerance
- Enhanced Oil Recovery Techniques
- Experimental Learning in Engineering
- Probabilistic and Robust Engineering Design
- Genome Rearrangement Algorithms
Robert Bosch (Germany)
2023-2024
RIKEN Center for Advanced Photonics
2023-2024
University of Maryland, College Park
2023-2024
ExxonMobil (United States)
2023-2024
RIKEN Center for Quantum Computing
2023-2024
Sandia National Laboratories
2023-2024
Poznan Supercomputing and Networking Center
2012-2023
Polish Academy of Sciences
2014-2019
Abstract Validation, verification, and uncertainty quantification (VVUQ) of simulation workflows are essential for building trust in results, their increased use decision‐making processes. The EasyVVUQ Python library is designed to facilitate implementation advanced VVUQ techniques new or existing workflows, with a particular focus on high‐performance computing, middleware agnosticism, multiscale modeling. Here, the application five very diverse areas demonstrated: materials properties,...
We present the VECMA toolkit (VECMAtk), a flexible software environment for single and multiscale simulations that introduces directly applicable reusable procedures verification, validation (V&V), sensitivity analysis (SA) uncertainty quantification (UQ). It enables users to verify key aspects of their applications, systematically compare validate simulation outputs against observational or benchmark data, run conveniently on any platform from desktop current multi-petascale computers. In...
Nature is observed at all scales; with multiscale modeling, scientists bring together several scales for a holistic analysis of phenomenon. The models on these different may require significant but also heterogeneous computational resources, creating the need distributed computing. A particularly demanding type models, tightly coupled, brings it number theoretical and practical issues. In this contribution, coupled model in-stent restenosis first theoretically examined its merits using...
We describe our Multiscale Computing Patterns software for High Performance Computing. Following a short review of Patterns, this paper introduces the Software, which consists description, optimisation and execution components. First, description component translates task graph, representing multiscale simulation, to particular type computing pattern. Second, selects applies algorithms find most suitable mapping between submodels available HPC resources. Third, middleware layer maps number...
Today scientists and engineers are commonly faced with the challenge of modelling, predicting controlling multiscale systems which cross scientific disciplines where several processes acting at different scales coexist interact. Such multidisciplinary models, when simulated in three dimensions, require large scale or even extreme computing capabilities. The MAPPER project is developing computational strategies, software services to enable distributed simulations across disciplines,...
With the recent advent of novel multi- and many-core hardware architectures, application programmers have to deal with many hardware-specific implementation details be familiar software optimization techniques benefit from new high-performance computing machines. Highly effcient parallel design is in fact an interdisciplinary process involving domain specific IT experts. Therefore, this paper aims present early experiences computationally demanding applications, development efforts...
We describe a method for queue wait time prediction in supercomputing clusters. It was designed use as part of multi-criteria brokering mechanisms resource selection multi-site High Performance Computing environment. The aim is to incorporate the jobs stay queued scheduling system into criteria. Our can also be used by end users estimate completion their computing jobs. uses historical data about particular make predictions. returns list probability estimates form ( t i, p i), where i that...
Not many fully integrated Grid solutions exist on today's market.In this paper we present the PSNC's toolkit, called Gridge, which is environment, consisting of tools and services to enable challenging scientific commercial applications Grid.
The growing demand for computational power causes that Grids are becoming mission-critical components in research and industry, offering sophisticated solutions leveraging large-scale computing storage resources.The nature a Grid which resources usually shared among multiple organizations under their control based on the "best effort" approach with no guarantee concerning quality-of-service may be inadequate to support simulations.Requirements of such simulations often exceed capabilities...
In large scale production and scientific, academic environments, the information sets to perform computations on come from various sources. particular, some may require obtained as a result of previous computations. Workflow description offers an attractive approach formally deal with such complex processes. Vine Toolkit solution addresses major challenges here synchronization distributed workflows, establishing community driven grid environment for seamless results sharing collaboration....