Computational Science : Scipedia

Migration of a generic multi-physics framework to HPC environments
P. Dadvand, R. Rossi, M. Gil, X. Martorell, J. Cotela, E. Juanpere, S. Idelsohn, E. Oñate

Computers and Fluids (2013). Vol. 80, pp. 301-309

Abstract
Creating a highly parallelizable code is a challenge specially for Distributed Memory Machines (DMMs). Moreover, algorithms and data structures suitable for these platforms can be very different from the ones used in serial code. For this reason, many programmers in the field prefer to start their own code from scratch. However, for an already existing framework supported by a long-time expertise the idea of transformation becomes attractive in order to reuse the effort done during years of development. In this presentation we explain how a relatively complex framework but with modular structure can be prepared for high performance computing with minimum modification. Kratos Multi-Physics is an open source generic multi-disciplinary platform for solution of coupled problems consist of fluid, structure, thermal and electromagnetic fields. The parallelization of this framework is performed with objective of enforcing the less possible changes to its different solver modules and encapsulate the changes as much as possible in its common kernel. This objective is achieved thanks to the Kratos design and also innovative way of dealing with data transfers for a multi-disciplinary code. This work is completed by the migration of the framework from the 86× architecture to the Marenostrum Supercomputing platform. The migration has been verified by a set of benchmarks which show high scalability, from which we present the Telescope problem in this paper.
Abstract
Creating a highly parallelizable code is a challenge specially for Distributed Memory Machines (DMMs). Moreover, algorithms and data structures suitable for these platforms [...]
- 19
- read
An Object-oriented Environment for Developing Finite Element Codes for Multi-disciplinary Applications
P. Dadvand, R. Rossi, E. Oñate

Archives of Comp. Meths. Engng. (2010). Vol. 17 (3), pp 253-297

Abstract
The objective of this work is to describe the design and implementation of a framework for building multi-disciplinary finite element programs. The main goals are generality, reusability, extendibility, good performance and memory efficiency. Another objective is preparing the code structure for team development to ensure the easy collaboration of experts in different fields in the development of multi-disciplinary applications. Kratos, the framework described in this work, contains several tools for the easy implementation of finite element applications and also provides a common platform for the natural interaction of different applications. To achieve this, an innovative variable base interface is designed and implemented. This interface is used at different levels of abstraction and showed to be very clear and extendible. A very efficient and flexible data structure and an extensible IO are created to overcome difficulties in dealing with multi-disciplinary problems. Several other concepts in existing works are also collected and adapted to coupled problems. The use of an interpreter, of different data layouts and variable number of dofs per node are examples of such approach. In order to minimize the possible conflicts arising in the development, a kernel and application approach is used. The code is structured in layers to reflect the working space of developers with different fields of expertise. Details are given on the approach chosen to increase performance and efficiency. Examples of application of Kratos to different multidisciplinary problems are presented in order to demonstrate the applicability and efficiency of the new object oriented environment.
Abstract
The objective of this work is to describe the design and implementation of a framework for building multi-disciplinary finite element programs. The main goals are generality, [...]
- 17
- read
Software Solutions for ICME
G. Schmitz, A. Engstrom, R. Bernhardt, U. Prahl, L. Adam, J. Seyfarth, M. Apel, C. Agelet de Saracibar, P. Korzhavyi, J. Agren, B. Patzak

Journal of The Minerals, Metals & Materials Society (2016). Vol. 68 (1), pp. 70-76

Abstract
The Integrated Computational Materials Engineering expert group (ICMEg), a coordination activity of the European Commission, aims at developing a global and open standard for information exchange between the heterogeneous varieties of numerous simulation tools. The ICMEg consortium coordinates respective developments by a strategy of networking stakeholders in the first International Workshop on Software Solutions for ICME, compiling identified and relevant software tools into the Handbook of Software Solutions for ICME, discussing strategies for interoperability between different software tools during a second (planned) international workshop, and eventually proposing a scheme for standardized information exchange in a future book or document. The present article summarizes these respective actions to provide the ICME community with some additional insights and resources from which to help move this field forward.
Abstract
The Integrated Computational Materials Engineering expert group (ICMEg), a coordination activity of the European Commission, aims at developing a global and open standard [...]
- 13
- read
Cache‐efficient renumbering for vectorization
R. Lohner

Int. J. Numer. Meth. Bio. Engng. (2008). Vol. 26 (5), pp. 628-636

Abstract
A renumbering strategy for field solvers based on unstructured grids that avoids memory contention and minimizes cache‐misses is described. Compared with usual colouring techniques, the new renumbering strategy reduces the spread in point‐data access for edge‐based solvers by more than an order of magnitude. The technique is particularly suited for multicore, cache‐based machines that allow for vectorization or pipelining.
Abstract
A renumbering strategy for field solvers based on unstructured grids that avoids memory contention and minimizes cache‐misses is described. Compared with usual colouring [...]
- 12
- read
Cache‐efficient renumbering for vectorization
R. Lohner

International Journal for Numerical Methods in Biomedical Engineering (2008). Vol. 26 (5), pp. 628-636

Abstract
A renumbering strategy for field solvers based on unstructured grids that avoids memory contention and minimizes cache‐misses is described. Compared with usual colouring techniques, the new renumbering strategy reduces the spread in point‐data access for edge‐based solvers by more than an order of magnitude. The technique is particularly suited for multicore, cache‐based machines that allow for vectorization or pipelining.
Abstract
A renumbering strategy for field solvers based on unstructured grids that avoids memory contention and minimizes cache‐misses is described. Compared with usual colouring [...]
- 4
- read
FEMPAR: an object-oriented parallel finite element framework
S. Badia, A. Martín, J. Principe

Archives of Comp. Meths. Engng. (2018). Vol. 25 (2), pp. 195-271

Abstract
FEMPAR is an open source object oriented Fortran200X scientific software library for the high-performance scalable simulation of complex multiphysics problems governed by partial differential equations at large scales, by exploiting state-of-the-art supercomputing resources. It is a highly modularized, flexible, and extensible library, that provides a set of modules that can be combined to carry out the different steps of the simulation pipeline. FEMPAR includes a rich set of algorithms for the discretization step, namely (arbitrary-order) grad, div, and curl-conforming finite element methods, discontinuous Galerkin methods, B-splines, and unfitted finite element techniques on cut cells, combined with h-adaptivity. The linear solver module relies on state-of-the-art bulk-asynchronous implementations of multilevel domain decomposition solvers for the different discretization alternatives and block-preconditioning techniques for multiphysics problems. FEMPAR is a framework that provides users with out-of-the-box state-of-the-art discretization techniques and highly scalable solvers for the simulation of complex applications, hiding the dramatic complexity of the underlying algorithms. But it is also a framework for researchers that want to experience with new algorithms and solvers, by providing a highly extensible framework. In this work, the first one in a series of articles about FEMPAR, we provide a detailed introduction to the software abstractions used in the discretization module and the related geometrical module. We also provide some ingredients about the assembly of linear systems arising from finite element discretizations, but the software design of complex scalable multilevel solvers is postponed to a subsequent work.
Abstract
FEMPAR is an open source object oriented Fortran200X scientific software library for the high-performance scalable simulation of complex multiphysics problems governed by [...]
- 3
- read
Implementation and Scalability Analysis of Balancing Domain Decomposition Methods
S. Badia, A. Martín, J. Principe

Archives of Comp. Meths. Engng. (2013). Vol. 20 (3), pp. 239-262

Abstract
In this paper we present a detailed description of a high-performance distributed-memory implementation of balancing domain decomposition preconditioning techniques. This coverage provides a pool of implementation hints and considerations that can be very useful for scientists that are willing to tackle large-scale distributed-memory machines using these methods. On the other hand, the paper includes a comprehensive performance and scalability study of the resulting codes when they are applied for the solution of the Poisson problem on a large-scale multicore-based distributed-memory machine with up to 4096 cores. Well-known theoretical results guarantee the optimality (algorithmic scalability) of these preconditioning techniques for weak scaling scenarios, as they are able to keep the condition number of the preconditioned operator bounded by a constant with fixed load per core and increasing number of cores. The experimental study presented in the paper complements this mathematical analysis and answers how far can these methods go in the number of cores and the scale of the problem to still be within reasonable ranges of efficiency on current distributed-memory machines. Besides, for those scenarios where poor scalability is expected, the study precisely identifies, quantifies and justifies which are the main sources of inefficiency.
Abstract
In this paper we present a detailed description of a high-performance distributed-memory implementation of balancing domain decomposition preconditioning techniques. This [...]
- 3
- read
Database Management in Scientific Computing Part I: General Description
C. Felippa

Computers and Structures (1980). Vol. 12, pp. 131-146

Abstract
Techniques for generalized data management are gradually making inroads in scientific computing in general and computerized engineering analysis in particular. This development parallels in many ways the rapid acceptance of the centralized database concept in business-oriented processing during the 1970s. Scientific database manipulation systems face, however, a specialized general features of scientific data management from a functional standpoint. Special emphasis is given to the discussion of trends in the development of large-scale programs, as well as the assembly of database-linked program networks for engineering analysis.
Abstract
Techniques for generalized data management are gradually making inroads in scientific computing in general and computerized engineering analysis in particular. This development [...]
- 3
- read
Balancing domain decomposition by constraints and perturbation
S. Badia, H. Nguyen

SIAM journal on numerical anaysis (2016). Vol. 54 (6), pp. 3436-3467

Abstract
In this paper, we formulate and analyze a perturbed formulation of the balancing domain decomposition by constraints (BDDC) method. We prove that the perturbed BDDC has the same polylogarithmic bound for the condition number as the standard formulation. Two types of properly scaled zero-order perturbations are considered: one uses a mass matrix, and the other uses a Robin-type boundary condition, i.e, a mass matrix on the interface. With perturbation, the wellposedness of the local Neumann problems and the global coarse problem is automatically guaranteed, and coarse degrees of freedom can be defined only for convergence purposes but not well-posedness. This allows a much simpler implementation as no complicated corner selection algorithm is needed. Minimal coarse spaces using only face or edge constraints can also be considered. They are very useful in extreme scale calculations where the coarse problem is usually the bottleneck that can jeopardize scalability. The perturbation also adds extra robustness as the perturbed formulation works even when the constraints fail to eliminate a small number of subdomain rigid body modes from the standard BDDC space. This is extremely important when solving problems on unstructured meshes partitioned by automatic graph partitioners since arbitrary disconnected subdomains are possible. Numerical results are provided to support the theoretical findings. -
Abstract
In this paper, we formulate and analyze a perturbed formulation of the balancing domain decomposition by constraints (BDDC) method. We prove that the perturbed BDDC [...]
- 2
- read
Nonlinear parallel-in-time Schur complement solvers for ordinary differential equations
S. Badia, M. Olm

Journal of Computational and Applied Mathematics (2018). Vol. 344, pp. 794-806

Abstract
In this work, we propose a parallel-in-time solver for linear and nonlinear ordinary differential equations. The approach is based on an efficient multilevel solver of the Schur complement related to a multilevel time partition. For linear problems, the scheme leads to a fast direct method. Next, two different strategies for solving nonlinear ODEs are proposed. First, we consider a Newton method over the global nonlinear ODE, using the multilevel Schur complement solver at every nonlinear iteration. Second, we state the global nonlinear problem in terms of the nonlinear Schur complement (at an arbitrary level), and perform nonlinear iterations over it. Numerical experiments show that the proposed schemes are weakly scalable, i.e., we can efficiently exploit increasing computational resources to solve for more time steps the same problem. -
Abstract
In this work, we propose a parallel-in-time solver for linear and nonlinear ordinary differential equations. The approach is based on an efficient multilevel solver [...]
- 2
- read
Fundamental Diagrams for Specific Very High Density Crowds
R. Lohner, B. Muhamad, P. Dambalmath, E. Haug

Collective Dynamics (2017). Vol. 2 (A13), pp.1-15

Abstract
An experimental campaign was undertaken to measure the pedestrian flow in the region close to the Kaaba during the Hajj pilgrimages of 2014 and 2015. High resolution video and photographs were used. The space was divided into areas of 10 sqm. The pedestrians were counted, and the velocity measured from video clips. The results were surprising: the velocity in the very high density region increases, which implies also an increase of the flux. The flux in this region (with more than 8 p/sqm) reaches values that exceed 3.5 p/m/sec, much higher than previously recorded under more `standard conditions' in corridors and passages.
Abstract
An experimental campaign was undertaken to measure the pedestrian flow in the region close to the Kaaba during the Hajj pilgrimages of 2014 and 2015. High resolution video [...]
- 2
- read
Simple fault-tolerant computing for field solvers
A. Degro, R. Lohner

Int. J. Numer. Meth. Fluids (2020). Vol. 34 (1), pp. 1-14

Abstract
Fault-tolerant computing options based on the use of restart information stored on and off node and the use of reserve processes have been developed, implemented and tested in a large-scale, production field solver taken from the domain of computational fluid dynamics. The tests conducted to date have shown good results, with recovery rates approaching 100% under realistic node failure scenarios. Even though the computational overhead of the field solver is very low (explicit time-marching and finite differences), the fault-tolerant implementation adds a run-time penalty that is only in the range of 6–12%, depending on the spatial and temporal approximation used. The procedures developed are generally applicable, and could easily be ported to other code.
Abstract
Fault-tolerant computing options based on the use of restart information stored on and off node and the use of reserve processes have been developed, implemented and tested [...]
- 2
- read
Parallelizing the construction of indirect access arrays for shared‐memory machines
J. Tuszynski, R. Lohner

Comput. Methods Appl. Mech. Engrg., (1998). Vol. 14 (8), pp. 773-781

Abstract
A way has been found to form indirect addressing lists in parallel on shared-memory parallel machines. The maximum possible speed-up for typical tetrahedral grids is approximately 1:23. The algorithm requires an additional scratch array to shift from the serial ‘elements surrounding points’ to the parallel ‘elements surrounding processors surrounding points’ paradigm. The algorithm developed is general in nature, i.e. applicable to all indirect addressing lists. All numerical methods requiring the construction of indirect data structures, such as sparse matrix linear algebra procedures, field and particle solvers operating on unstructured grids, and network flow applications should see a benefit from this algorithm when running on shared-memory parallel machines
Abstract
A way has been found to form indirect addressing lists in parallel on shared-memory parallel machines. The maximum possible speed-up for typical tetrahedral grids is approximately [...]
- 2
- read
Multilevel balancing domain decomposition at extreme scales
S. Badia, A. Martín, J. Principe

SIAM J. Numer. Anal. (2016). (preprint) Vol. 38 (1), pp. C22-C52

Abstract
In this paper we present a fully distributed, communicator-aware, recursive, and interlevel-overlapped message-passing implementation of the multilevel balancing domain decomposition by constraints (MLBDDC) preconditioner. The implementation highly relies on subcommunicators in order to achieve the desired effect of coarse-grain overlapping of computation and communication, and communication and communication among levels in the hierarchy (namely, interlevel overlapping). Essentially, the main communicator is split into as many nonoverlapping subsets of message-passing interface (MPI) tasks (i.e., MPI subcommunicators) as levels in the hierarchy. Provided that specialized resources (cores and memory) are devoted to each level, a careful rescheduling and mapping of all the computations and communications in the algorithm lets a high degree of overlapping be exploited among levels. All subroutines and associated data structures are expressed recursively, and therefore MLBDDC preconditioners with an arbitrary number of levels can be built while re-using significant and recurrent parts of the codes. This approach leads to excellent weak scalability results as soon as level-1 tasks can fully overlap coarser-levels duties. We provide a model to indicate how to choose the number of levels and coarsening ratios between consecutive levels and determine qualitatively the scalability limits for a given choice. We have carried out a comprehensive weak scalability analysis of the proposed implementation for the three-dimensional Laplacian and linear elasticity problems on structured and unstructured meshes. Excellent weak scalability results have been obtained up to 458,752 IBM BG/Q cores and 1.8 million MPI being, being the first time that exact domain decomposition preconditioners (only based on sparse direct solvers) reach these scales. (An erratum is attached.)
Abstract
In this paper we present a fully distributed, communicator-aware, recursive, and interlevel-overlapped message-passing implementation of the multilevel balancing domain decomposition [...]
- 1
- read
Docencia e investigación: ¿un falso dilema?
E. Montilla, G. Amblàs, G. Bugeda, S. Roca

Aula abierta (2013). Vol. 41 (2), pp. 13-22

Abstract
La Universidad Politécnica de Cataluña (UPC) dispone de un modelo de evaluación docente plenamente consolidado, que está adaptado a los requerimientos que emanan de la implantación del Espacio Europeo de Educación Superior. Dicho modelo de evaluación ha sido homologado por la Agència per a la Qualitat del Sistema Universitari de Catalunya (AQU Catalunya), la agencia de evaluación catalana y se basa en las directrices europeas más exigentes. Este modelo se ha empleado para evaluar la docencia del profesorado en seis convocatorias, por lo que se dispone de una información exhaustiva, tanto a nivel agregado como separado por colectivos y ámbitos de conocimiento. Por otro lado, también se dispone de información sobre la actividad investigadora del profesorado, medida mediante el número de sexenios obtenidos. En este artículo se presentan estos datos y se estudia si existe una correlación positiva entre la calidad de la docencia impartida y el reconocimiento de las tareas investigadoras de los profesores, medida a través del número de sexenios obtenidos. Nuestros resultados indican que tal correlación existe, aunque es más débil de lo que cabría esperar.
Abstract
La Universidad Politécnica de Cataluña (UPC) dispone de un modelo de evaluación docente plenamente consolidado, que está adaptado a los requerimientos [...]
- 1
- read
On maximum achievable speeds for field solvers
R. Lohner, J. Baum

Int. J. Numer. Meth. Heat and Fluid Flows (2014). Vol. 24 (7), pp. 1537-1544

Abstract
Purpose ‐ Prompted by the empirical evidence that achievable flow solver speeds for large problems are limited by what appears to be a time of the order of O(0.1) sec/timestep regardless of the number of cores used, the purpose of this paper is to identify why this phenomenon occurs. Design/methodology/approach ‐ A series of timing studies, as well as in-depth analysis of memory and inter-processors transfer requirements were carried out for a typical field solver. The results were analyzed and compared to the expected performance. Findings ‐ The analysis shows that at present flow speeds per core are already limited by the achievable transfer rate to RAM. For smaller domains/larger number of processors, the limiting speed of CFD solvers is given by the MPI communication network. Research limitations/implications ‐ This implies that at present, there is a "limiting useful size" for domains, and that there is a lower limit for the time it takes to update a flowfield. Practical implications ‐ For practical calculations this implies that the time required for running large-scale problems will not decrease markedly once these applications migrate to machines with hundreds of thousands of cores. Originality/value ‐ This is the first time such a finding has been reported in this context.
Abstract
Purpose ‐ Prompted by the empirical evidence that achievable flow solver speeds for large problems are limited by what appears to be a time of the order of O(0.1) sec/timestep [...]
- 1
- read
Renumbering strategies for unstructured-grid solvers operating on shared-memory, cache-based parallel machines
R. Lohner

Comput. Methods Appl. Mech. Engrg., (1998). Vol. 163 (1-4), pp. 95-109

Abstract
Two renumbering strategies for field solvers based on unstructured grids that operate on shared-memory, cache-based parallel machines are described. Special attention is paid to the avoidance of cache-line overwrite, which can lead to drastic performance degradation on this type of machines. Both renumbering techniques avoid cache-misses and cache-line overwrite while allowing pipelining, leading to optimal coding for this type of hardware
Abstract
Two renumbering strategies for field solvers based on unstructured grids that operate on shared-memory, cache-based parallel machines are described. Special attention is paid [...]
- 1
- read
Everyone may benefit from subsidising entry to risky occupations
C. Felippa, K. Park, J. Black, D. Meza

Journal of Public Economics (1997). Vol. 66 (3), pp. 409-424

Abstract
This paper shows that in the presence of costly state verification, directly or indirectly subsidising entry to risky occupations may benefit everyone. The result holds even in the presence of private insurance. Indeed, it may be desirable to prohibit private insurance in favour of subsidies to hazardous activities. These findings do not depend on the government having an advantage over the private sector in observing outcomes. The explanation is that through its influence on equilibrium price, feasible fiscal policy can shift the return distribution so as to create collective insurance more cheaply than is possible through private contracting with its requirement of costly auditing. Amongst applications is a case for a loss-making state bank offering high interest-rate loans.
Abstract
This paper shows that in the presence of costly state verification, directly or indirectly subsidising entry to risky occupations may benefit everyone. The result holds even [...]
- 1
- read
Database management in scientific computing—II. Data structures and program architecture
C. Felippa

Computers and Structures (1980). Vol. 12 (1), pp. 131-145

Abstract
The second of a three-part paper on scientific database management prepares the ground for the operational description of a multilevel data management system in Part III. Following a detailed categorization of data and storage structures, principles of program architecture deemed necessary to make most efficient use of centralized data management techniques are discussed. General operational requirements of data management systems are presented, and the notion of activity levels introduced. This part closes with a brief review of the evolution of data management techniques in the context of computerized structural analysis.
Abstract
The second of a three-part paper on scientific database management prepares the ground for the operational description of a multilevel data management system in Part III. [...]
- 0
- read
Professional workstations
C. Felippa

Engineering Computations (1984). Vol. 1 (2), pp. 183-188

Abstract
Professional workstations are dedicated computers that possess sufficient hardware power and software capabilities to allow the routine performance of job-related activities by professional users. The advent of these computers reflects current trends in decentralization and specialization of computer equipment. The paper presents an informal review of classes of workstations of potential interest to engineers and scientists. The review was motivated by projected computing needs in a software- intensive research-and-development industrial laboratory.
Abstract
Professional workstations are dedicated computers that possess sufficient hardware power and software capabilities to allow the routine performance of job-related activities [...]
- 0
- read
A command reader for interactive programming
C. Felippa

Engineering Computations (1985). Vol. 2 (3), pp. 203-237

Abstract
A simplified command line reader suitable for building interactive programs based on Fortran 77 is presented. The reader is a 20:1 condensation of the comprehensive command language interpreter CLIP that supports the NICE integrated software system. The present reader, called TinyCLIP. has been prepared to illustrate basic elements of command-driven programming for beginners interested in writing Fortran-based interactive applications software. The reader is table driven. Data lines given to the reader are parsed into items which are stored in a table. The table can be subsequently accessed by the program that processes the commands. Tools for extending the basic language are briefly discussed and specific extensions suggested.
Abstract
A simplified command line reader suitable for building interactive programs based on Fortran 77 is presented. The reader is a 20:1 condensation of the comprehensive command [...]
- 0
- read
Towards parallel I/O in finite element simulations
C. Farhat, E. Pramono, C. Felippa

International Journal for Numerical Methods in Engineering (1989). Vol. 28 (11)

Abstract
I/O issues in finite element analysis on parallel processors are addressed. Viable solutions for both local and shared memory multiprocessors are presented. The approach is simple but limited by currently available hardware and software systems. Implementation is carried out on a CRAY-2 system. Performance results are reported.
Abstract
I/O issues in finite element analysis on parallel processors are addressed. Viable solutions for both local and shared memory multiprocessors are presented. The approach is [...]
- 0
- read
Handling tens of thousands of cores with industrial/legacy codes: Approaches, implementation and timings
R. Lohner, J. Baum

Computers and Fluids (2013). Vol. 85, pp. 53-62

Abstract
The consequences that the recent stagnation in clockrates for CPUs has had on large-scale CFD runs are examined. At first sight, the conclusion is that only massive parallelism at the loop or domain decomposition level will lead to higher FLOP counts. However, the significant differences in advances for CPUs/GPUs versus RAM and interprocessor communication bandwidth lead to a so-called ‘limiting domain size’, below which communication dominates execution times and performance degrades drastically. The consequences of this ‘red-shift’ for the future of CFD are manifold: the time to advance the solution one timestep is limited, implying that even with unlimited number of processors/cores, LES, DES and DNS runs for realistic Reynolds-numbers will require days or weeks of execution.
Abstract
The consequences that the recent stagnation in clockrates for CPUs has had on large-scale CFD runs are examined. At first sight, the conclusion is that only massive parallelism [...]
- 0
- read

information

About this publication

How to submit

Open access

Contact

Migration of a generic multi-physics framework to HPC environments

An Object-oriented Environment for Developing Finite Element Codes for Multi-disciplinary Applications

Software Solutions for ICME

Cache‐efficient renumbering for vectorization

Cache‐efficient renumbering for vectorization

FEMPAR: an object-oriented parallel finite element framework

Implementation and Scalability Analysis of Balancing Domain Decomposition Methods

Database Management in Scientific Computing Part I: General Description

Balancing domain decomposition by constraints and perturbation

Nonlinear parallel-in-time Schur complement solvers for ordinary differential equations

Fundamental Diagrams for Specific Very High Density Crowds

Simple fault-tolerant computing for field solvers

Parallelizing the construction of indirect access arrays for shared‐memory machines

Multilevel balancing domain decomposition at extreme scales

Docencia e investigación: ¿un falso dilema?

On maximum achievable speeds for field solvers

Renumbering strategies for unstructured-grid solvers operating on shared-memory, cache-based parallel machines

Everyone may benefit from subsidising entry to risky occupations

Database management in scientific computing—II. Data structures and program architecture

Professional workstations

A command reader for interactive programming

Towards parallel I/O in finite element simulations

Handling tens of thousands of cores with industrial/legacy codes: Approaches, implementation and timings

information

Sign up at Scipedia and complete your profile

Click on "CREATE A DOCUMENT" in your profile page

Upload your manuscript in LaTeX or Word, or create it from scratch with our online editor

Click on "SUBMIT FOR PUBLICATION" at the manuscript page, and select your preferred journal or collection

Click on "SUBMIT FOR PUBLICATION" at the manuscript page, select your preferred collection and complete the submission form

The editorial board will ensure that the publishing requirements are adhered to, before your paper is published