A5016308631- Parallel Computing and Optimization Techniques
- Distributed and Parallel Computing Systems
- Advanced Data Compression Techniques
- Meteorological Phenomena and Simulations
- Precipitation Measurement and Analysis
- Advanced Numerical Methods in Computational Mathematics
- Computational Physics and Python Applications
- Matrix Theory and Algorithms
- Cloud Computing and Resource Management
- Digital Filter Design and Implementation
- Algorithms and Data Compression
- Advanced Data Storage Technologies
- Multimedia Communication and Technology
- Image and Signal Denoising Methods
- Numerical methods for differential equations
- Computational Geometry and Mesh Generation
- Scientific Computing and Data Management
- Computer Graphics and Visualization Techniques
- BIM and Construction Integration
- Geophysics and Gravity Measurements
- 3D Shape Modeling and Analysis
- Gas Dynamics and Kinetic Theory
- Image Retrieval and Classification Techniques
- Interconnection Networks and Systems
- Computational Fluid Dynamics and Aerodynamics
Poznan Supercomputing and Networking Center
2011-2020
Poznań University of Technology
2014-2016
Abstract. In the simulation of complex multi-scale flows arising in weather and climate modelling, one biggest challenges is to satisfy strict service requirements terms time solution budgetary constraints energy solution, without compromising accuracy stability application. These simulations require algorithms that minimise footprint along with required produce a maintain physically level accuracy, are numerically stable, resilient case hardware failure. The European Centre for Medium-Range...
Summary The goal of this study is to adapt the multiscale fluid solver EULerian or LAGrangian framewrok (EULAG) future graphics processing units (GPU) platforms. EULAG model has proven record successful applications, and excellent efficiency scalability on conventional supercomputer architectures. Currently, being implemented as new dynamical core COSMO weather prediction framework. Within study, two main modules EULAG, namely multidimensional positive definite advection transport algorithm...
Hyperspectral image compression has received considerable interest in recent years due to the enormous data volumes collected by imaging spectrometers for Earth Observation. JPEG2000 is an important technique compression, which been successfully used context of hyperspectral either lossless and lossy fashion. Due increasing spatial, spectral, temporal resolution remotely sensed sets, fast (on-board) becoming challenging objective, with potential reduce limitations downlink connection between...
Starting from a high-level problem description in terms of partial differential equations using abstract tensor notation, the Chemora framework discretizes, optimizes, and generates complete high performance codes for wide range compute architectures. extends capabilities Cactus, facilitating usage large-scale CPU/GPU systems an efficient manner complex applications, without low-level code tuning. achieves parallelism through MPI multi-threading, combining OpenMP CUDA. Optimizations include...
The use of graphics hardware for non-graphics applications has become popular among many scientific programmers and researchers as we have observed a higher rate theoretical performance increase than the CPUs in recent years. However, gains may be easily lost context specific parallel application due to various both software factors. Consequently, benchmarks testing are still best techniques compare effciency emerging architectures with built-in support parallelism at different levels....
Hyperspectral image compression has received considerable interest in recent years due to the enormous data volumes collected by imaging spectrometers for Earth Observation. JPEG2000 is an important technique which been successfully used context of hyperspectral compression, either lossless and lossy fashion. Due increasing spatial, spectral temporal resolution remotely sensed sets, fast (onboard) becoming a very challenging objective, with potential reduce limitations downlink connection...
Performance of high-end supercomputers will reach the exascale through advent core counts in billions. However, upcoming computing era it is important not only to focus on performance, but also scalability fine-grained parallel applications, data locality and energy aware scheduling within code. In fact, applications need change even now by redesigning algorithms structures respectively take advantage recent improvements efficiency heterogeneous hardware, including multicore processors GPU...
Abstract. In the simulation of complex multi-scale flow problems, such as those arising in weather and climate modelling, one biggest challenges is to satisfy operational requirements terms time-to-solution energy-to-solution yet without compromising accuracy stability calculation. These competing factors require development state-of-the-art algorithms that can optimally exploit targeted underlying hardware efficiently deliver extreme computational capabilities typically required forecast...
The recent advent of novel multi- and many-core architectures forces application programmers to deal with hardware-specific implementation details be familiar software optimisation techniques benefit from new high-performance computing machines. Extra care must taken for communication-intensive algorithms, which may a bottleneck forthcoming era exascale computing. This paper aims present high-level stencil framework implemented the EULerian or LAGrangian model (EULAG) that efficiently...
Starting from a high-level problem description in terms of partial differential equations using abstract tensor notation, the Chemora framework discretizes, optimizes, and generates complete high performance codes for wide range compute architectures. extends capabilities Cactus, facilitating usage large-scale CPU/GPU systems an efficient manner complex applications, without low-level code tuning. achieves parallelism through MPI multi-threading, combining OpenMP CUDA. Optimizations include...
Abstract Heterogeneous many-core computing resources are increasingly popular among users due to their improved performance over homogeneous systems. Many developers have realized that heterogeneous systems, e.g. a combination of shared memory multi-core CPU machine with massively parallel Graphics Processing Units (GPUs), can provide significant opportunities wide range applications. However, the best overall only be achieved if application tasks efficiently assigned different types...
Poznan Supercomputing and Networking Services (PSNC) offers advanced visualisation multimedia infrastructures as a set of dedicated laboratories to conduct innovative research development projects involving both academia industry. In this short overview we present the existing facilities located at PSNC campus in Poznan, Poland well descriptions example applications networked services which have been recently developed.