Computer Science
Loading...
ISSN 1508-2806
e-ISSN: 2300-7036
Call number
Volume
Vol. 12
Date
2011
Description
Reviewed by: Bipin Indurkhya, Joanna Grabska-Chrząstowska, Michał Turała, Renata Słota, Krzysztof Cetnarowicz, Marian Bubak, Mariusz Sterzel, Paweł Russek, Andrzej Ruta, Ernest Jamro, Kazimierz Wiatr, Jacek Kitowski, Łukasz Dutka, Jacek Rokicki, Piotr Breitkopf, Witold Cecot, Krzysztof Banaś, Włodzimierz Funika, Bartosz Kryza
Journal
Computer Science
AGH University Press (2004-)
ISSN: 1508-2806 e-ISSN: 2300-7036
ISSN: 1508-2806 e-ISSN: 2300-7036
journal.volume.project
Contains
Journal Issues
Articles
Evaluating new architectural features of the Intel(r) Xeon(r) 7500 processor for HPC workloads
(Wydawnictwa AGH, 2011) Gepner, Paweł; Fraser, David L.; Kowalik, Michał Filip; Waćkowski, Kazimierz
In this paper we take a look at what the Intel Xeon Processor 7500 family, code named Nehalem-EX, brings to high performance computing. We compare two families of Intel Xeon based systems (Intel Xeon 7500 and Intel Xeon 5600) and present a performance evolution of 16 node clusters based on these CPUs. We compare CPU generations utilizing dual socket platforms and a cluster across a number of HPC benchmarks and focused on different performance field and aspect. We will evaluate also technologies and features like Intels Hyper Threading Technology (HT) and Intel Turbo Boost Technology (Turbo Mode) and the performance implication of these technologies for HPC.
Benchmarking high performance architectures with natural language processing algorithms
(Wydawnictwa AGH, 2011) Kuta, Marcin; Kitowski, Jacek
Natural Language Processing algorithms are resource demanding, especially when tuning to inflective language like Polish is needed. The paper presents time and memory requirements of part of speech tagging and clustering algorithms applied to two corpora of the Polish language. The algorithms are benchmarked on three high performance platforms of different architectures. Additionally sequential versions and OpenMP implementations of clustering algorithms were compared.
Grid resource registry - abstract layer for computational resources
(Wydawnictwa AGH, 2011) Kasztelnik, Marek; Bubak, Marian
The growing number of resources available to researchers in the e-Science domain has opened new possibilities for constructing complex scientific applications while at the same time introducing new requirements for tools which assist developers in creating such applications. This paper discusses the problems of rapid application development, the use of distributed resources and a uniform approach to resource registration, discovery and access. It presents the Grid Resource Registry, which delivers an abstract layer for computational resources. The Registry is a central place where developers may search for available services and from which the execution engine receives technical specifications of services. The Registry is used throughout the lifetime ofthe e-science application, starting with application design, through implementation to execution.
Towards a grid infrastructure for hydro-meteorological research
(Wydawnictwa AGH, 2011) Schiffers, Michael; Kranzlmüller, Dieter; Clematis, Andrea; D'Agostino, Daniele; Galizia, Antonella; Quarati, Alfonso; Parodi, Antonio; Morando, Marina; Rebora, Nicola; Trasforini, Eva; Molini, Luca; Siccardi, Franko; Craig, George C.; Tafferner, Arnold
The Distributed Research Infrastructure for Hydro-Meteorological Study (DRIHMS) is a co-ordinated action co-funded by the European Commission. DRIHMS analyzes the main issues that arise when designing and setting up a pan-European Grid-based e-Infrastructure for research activities in the hydrologic and meteorological fields. The main outcome of the project is represented first by a set of Grid usage patterns to support innovative hydro-meteorological research activities, and second by the implications that such patterns define for a dedicated Grid infrastructure and the respective Grid architecture.
Convergence of iterative solvers for non-linear step-and-flash imprint lithography simulations
(Wydawnictwa AGH, 2011) Paszyński, Maciej
The paper presents the analysis of the iterative solvers utilized to solve the non-linear problem of Step-and-Flash Imprint Lithography (SFIL) a modern patterning process. The simulations consists in solving molecular statics problem for the polymer network, with quadratic potentials. The model distinguishes the strong interparticle interactions between particles forming a polymer network, and weak interactions between remaining particles. It also allows for large deformations, which all together implies the non-linear model. To illustrate the convergence of the iterative solvers, we present snapshots of the deformation of the sample being subject to the iterative solution. We claim that the animation is an interesting way of illustrating the convergence of the iterative solvers.