508 research outputs found
Recensiones [Revista de Historia Económica Año XVIII Otoño-Invierno 2000 n. 3 pp. 687-734]
Get PDFEditada en la Fundación Empresa PúblicaColl, S., y Guijarro, M.: Estadística aplicada a las Ciencias Sociales (Por Daniel Peña).-- Tedde de Lorca, P.: El Banco de San Femando (1829-1856) (Por Carlos Marichal).-- Comín Comín, F., y Martín Aceña, P.: Tabacalera y el estanco de tabaco en España (1636-1998) (Por Lina Gálvez Muñoz).-- Millán García-Várela, J.: El poder de la tierra. La sociedad agraria del bajo Segura en la época del liberalismo (Por Ricardo Robledo).-- Matés Barco, J. M.: La conquista del agua. Historia económica del abastecimiento urbano (Por Beatriz Mera González).-- Ortiz Batalla, J.: Los Bancos Centrales en América Latina (Por Raúl García Heras).-- Guirao, F.: Spain and the Reconstruction of Western Europe, 1945-57: Challenge and Response (Por Jordi Catalán).-- Aghion, P., y Howitt, P.: Endogenous Growth Theory (Por Joan R. Rosés).-- Dye, A. D.: Cuban Sugar in the Age of Mass Production. Technology and the Economics of Sugar Central, 1899-1929 (Por Antonio Santamaría García).-- Gourvish, T. R., y Tiratsoo, N. (eds.): Missionaries and managers: American influences on European management education, 1945-60 (Por Nuria Puig).-- Coastworth, J., y Taylor, A. (eds.): Latin America and the World Economy since 1800 (Por Gustavo A. del Ángel-Mobarak)Publicad
Resilience of Parallel Applications
Get PDFProceedings of the First PhD Symposium on Sustainable Ultrascale
Computing Systems (NESUS PhD 2016) Timisoara, Romania. February 8-11, 2016.Future exascale systems are predicted to be formed by millions of cores. This is a great opportunity for HPC
applications, however, it is also a hazard for the completion of their execution. Even if one computation node
presents a failure every one century, a machine with 100.000 nodes will encounter a failure every 9 hours. Thus,
HPC applications need to make use of fault tolerance techniques to ensure they successfully finish their execution.
This PhD thesis is focused on fault tolerance solutions for generic parallel applications, more specifically in checkpointing
solutions. We have extended CPPC, an MPI application-level portable checkpointing tool developed in
our research group, to work with OpenMP applications, and hybrid MPI-OpenMP applications. Currently, we
are working on transparently obtaining resilient MPI applications, that is, applications that are able to recover
themselves from failures without stopping their execution.European Cooperation in Science and Technology. COSTThis research was supported by the Ministry of Economy and Competitiveness of Spain and FEDER funds
of the EU (Project TIN2013-42148-P, and the predoctoral grant of Nuria Losada ref. BES-2014-068066) and
by EU under the COST Program Action IC1305: Network for Sustainable Ultrascale Computing (NESUS)
First person – Maria Losada-Perez
No full textFirst Person is a series of interviews with the first authors of a selection of papers published in Disease Models & Mechanisms, helping early-career researchers promote themselves alongside their papers. Maria Losada-Perez is first author on ‘
A novel injury paradigm in the central nervous system of adult Drosophila: Molecular, cellular and functional aspects’, published in DMM. Maria is a postdoc in the lab of Sergio Casas-Tintó at the Cajal Institute (CSIC) in Madrid, Spain, investigating glial cell responses to central nervous system (CNS) injuries and, in this context, cell communication
A Portable and Adaptable Fault Tolerance Solution for Heterogeneous Applications
Get PDF[Abstract] Heterogeneous systems have increased their popularity in recent years due to the high performance and reduced energy consumption capabilities provided by using devices such as GPUs or Xeon Phi accelerators. This paper proposes a checkpoint-based fault tolerance solution for heterogeneous applications, allowing them to survive fail-stop failures in the host CPU or in any of the accelerators used. Besides, applications can be restarted changing the host CPU and/or the accelerator device architecture, and adapting the computation to the number of devices available during recovery. The proposed solution is built combining CPPC (ComPiler for Portable Checkpointing), an application-level checkpointing tool, and HPL (Heterogeneous Programming Library), a library that facilitates the development of OpenCL-based applications. Experimental results show the low overhead introduced by the proposal and prove its portability and adaptability benefits.This research was supported by the Ministry of Economy and Competitiveness of Spain and FEDER funds of the EU (Projects TIN2013-42148-P, TIN2016-75845-P and the predoctoral Grant of Nuria Losada Ref. BES-2014-068066), by EU under the COST Program Action IC1305, Network for Sustainable Ultrascale Computing (NESUS), and by the Galician Government (Xunta de Galicia) and FEDER funds of the EU under the Consolidation Program of Competitive Research (Ref. GRC2013/055)Xunta de Galicia; GRC 2013/05
Towards Ad Hoc Recovery for Soft Errors
No full textThe coming exascale era is a great opportunity for high performance computing (HPC) applications. However, high failure rates on these systems will hazard the successful completion of their execution. Bit-flip errors in dynamic random access memory (DRAM) account for a noticeable share of the failures in supercomputers. Hardware mechanisms, such as error correcting code (ECC), can detect and correct single-bit errors and can detect some multi-bit errors while others can go undiscovered. Unfortunately, detected multi-bit errors will most of the time force the termination of the application and lead to a global restart. Thus, other strategies at the software level are needed to tolerate these type of faults more efficiently and to avoid a global restart. In this work, we extend the FTI checkpointing library to facilitate the implementation of custom recovery strategies for MPI applications, minimizing the overhead introduced when coping with soft errors. The new functionalities are evaluated by implementing local forward recovery on three HPC benchmarks with different reliability requirements. Our results demonstrate a reduction on the recovery times by up to 14%.This project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie grant agreement No 708566 (DURO). This research is also supported by the Ministry of Economy and Competitiveness of Spain and FEDER funds of
the EU (Projects TIN2016-75845-P and the predoctoral grant of Nuria Losada ref. BES-2014-068066), and by the Galician
Government (Xunta de Galicia) under the Consolidation Program of Competitive Research (ref. ED431C 2017/04).Peer Reviewe
Towards Ad Hoc Recovery for Soft Errors
Get PDFThe coming exascale era is a great opportunity for high performance computing (HPC) applications. However, high failure rates on these systems will hazard the successful completion of their execution. Bit-flip errors in dynamic random access memory (DRAM) account for a noticeable share of the failures in supercomputers. Hardware mechanisms, such as error correcting code (ECC), can detect and correct single-bit errors and can detect some multi-bit errors while others can go undiscovered. Unfortunately, detected multi-bit errors will most of the time force the termination of the application and lead to a global restart. Thus, other strategies at the software level are needed to tolerate these type of faults more efficiently and to avoid a global restart. In this work, we extend the FTI checkpointing library to facilitate the implementation of custom recovery strategies for MPI applications, minimizing the overhead introduced when coping with soft errors. The new functionalities are evaluated by implementing local forward recovery on three HPC benchmarks with different reliability requirements. Our results demonstrate a reduction on the recovery times by up to 14%.This project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie grant agreement No 708566 (DURO). This research is also supported by the Ministry of Economy and Competitiveness of Spain and FEDER funds of
the EU (Projects TIN2016-75845-P and the predoctoral grant of Nuria Losada ref. BES-2014-068066), and by the Galician
Government (Xunta de Galicia) under the Consolidation Program of Competitive Research (ref. ED431C 2017/04).Peer ReviewedPostprint (author's final draft
Addressing the challenges of climate change risks and adaptation in coastal areas: A review
No full textClimate change is and will continue altering the world's coasts, which are the most densely populated and economically active areas on earth and home for highly valuable ecosystems. While there is considerable relevant research, in the authors' experience this problem remains challenging for coastal engineering. This paper reviews important challenges in this respect and identifies three key actions to address them: (a) refocusing traditional practice towards more climate-aware approaches; (b) developing more comprehensive risk frameworks that include the multi-dimensionality and non-stationarity of their components and consideration of uncertainty; and (c) building bridges between risk assessment and adaptation theory and practice. We conclude that the way forward includes numerous activities including increased observations; the attribution of coastal impacts to their drivers; enhanced climate projections and their integration into impact models; more impact assessments at the local scale; dynamic projections of spatially-distributed exposure and vulnerability; and the exploration of inherently adaptive options. Given the complexity of the possible solutions, more practical guidance is required.Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.Coastal Engineerin
El bienestar escolar desde la experiencia del alumnado de Educación Primaria
Get PDFLa valoración de la experiencia del estudiante en la escuela a través de sus juicios cognitivos y sus afectos, o su percepción sobre su funcionamiento positivo y sus relaciones con otros influyen en su bienestar escolar. El objetivo de este estudio fue conocer los aspectos que producen bienestar escolar al alumnado. Se realizó un estudio cualitativo con enfoque fenomenológico para indagar en la experiencia de 302 estudiantes de Educación Primaria que respondieron al Cuestionario Abierto de Bienestar Escolar (CABES). La información se analizó inductivamente mediante un proceso cíclico obteniendo cuatro categorías: proceso de enseñanza-aprendizaje, espacios y tiempos escolares, relaciones y emociones. Destacó la relevancia otorgada a las relaciones entre iguales y con el profesorado, la demanda de mejoras en la enseñanza (aprendizaje cooperativo, basado en el juego), la gestión del tiempo (de estudio, libre) y el acondicionamiento del centro y del aula. Primaron las emociones positivas frente a las negativas, sobre todo la felicidad y la alegría, aunque también aparecieron frecuentemente aburrimiento y tristeza. Este estudio contribuirá a construir un instrumento para medir el bienestar escolar a partir de las categorías y de los códigos basados en la valoración cognitiva y afectiva de las experiencias del alumnado durante su jornada escolar
Local Rollback for Resilient Mpi Applications With Application-Level Checkpointing and Message Logging
Get PDF[Abstract]
The resilience approach generally used in high-performance computing (HPC) relies on coordinated checkpoint/restart, a global rollback of all the processes that are running the application. However, in many instances, the failure has a more localized scope and its impact is usually restricted to a subset of the resources being used. Thus, a global rollback would result in unnecessary overhead and energy consumption, since all processes, including those unaffected by the failure, discard their state and roll back to the last checkpoint to repeat computations that were already done. The User Level Failure Mitigation (ULFM) interface – the last proposal for the inclusion of resilience features in the Message Passing Interface (MPI) standard – enables the deployment of more flexible recovery strategies, including localized recovery. This work proposes a local rollback approach that can be generally applied to Single Program, Multiple Data (SPMD) applications by combining ULFM, the ComPiler for Portable Checkpointing (CPPC) tool, and the Open MPI VProtocol system-level message logging component. Only failed processes are recovered from the last checkpoint, while consistency before further progress in the execution is achieved through a two-level message logging process. To further optimize this approach point-to-point communications are logged by the Open MPI VProtocol component, while collective communications are optimally logged at the application level—thereby decoupling the logging protocol from the particular collective implementation. This spatially coordinated protocol applied by CPPC reduces the log size, the log memory requirements and overall the resilience impact on the applications.This research was supported by the Ministry of Economy and Competitiveness of Spain and FEDER funds of the EU (Projects TIN2016-75845-P and the predoctoral grants of Nuria Losada ref. BES-2014-068066 and ref. EEBB-I-17-12005); by EU under the COST Program Action IC1305 Network for Sustainable Ultrascale Computing (NESUS) and a HiPEAC Collaboration Grant and by the Galician Government (Xunta de Galicia) under the Consolidation Program of Competitive Research (ref. ED431C 2017/04). We gratefully thank Galicia Supercomputing Center for providing access to the FinisTerrae-II supercomputer.
This material is also based upon work supported by the US National Science Foundation, Office of Advanced Cyberinfrastructure , under Grants No. #1664142 and #1339763Xunta de Galicia; ED431C 2017/04US National Science Foundation, Office of Advanced Cyberinfrastructure; 1664142US National Science Foundation, Office of Advanced Cyberinfrastructure; 133976
Application-level Fault Tolerance and Resilience in HPC Applications
Get PDF[Resumo]
As necesidades computacionais das distintas ramas da ciencia medraron enormemente
nos últimos anos, o que provocou un gran crecemento no rendemento proporcionado
polos supercomputadores. Cada vez constrúense sistemas de computación
de altas prestacións de maior tamaño, con máis recursos hardware de distintos tipos,
o que fai que as taxas de fallo destes sistemas tamén medren. Polo tanto, o
estudo de técnicas de tolerancia a fallos eficientes é indispensábel para garantires
que os programas científicos poidan completar a súa execución, evitando ademais
que se dispare o consumo de enerxía. O checkpoint/restart é unha das técnicas máis
populares. Sen embargo, a maioría da investigación levada a cabo nas últimas décadas
céntrase en estratexias stop-and-restart para aplicacións de memoria distribuída
tralo acontecemento dun fallo-parada. Esta tese propón técnicas checkpoint/restart
a nivel de aplicación para os modelos de programación paralela roáis populares en
supercomputación. Implementáronse protocolos de checkpointing para aplicacións
híbridas MPI-OpenMP e aplicacións heteroxéneas baseadas en OpenCL, en ámbolos
dous casos prestando especial coidado á portabilidade e maleabilidade da solución.
En canto a aplicacións de memoria distribuída, proponse unha solución de resiliencia
que pode ser empregada de forma xenérica en aplicacións MPI SPMD, permitindo
detectar e reaccionar a fallos-parada sen abortar a execución. Neste caso, os procesos
fallidos vólvense a lanzar e o estado da aplicación recupérase cunha volta atrás global.
A maiores, esta solución de resiliencia optimizouse implementando unha volta
atrás local, na que só os procesos fallidos volven atrás, empregando un protocolo de
almacenaxe de mensaxes para garantires a consistencia e o progreso da execución.
Por último, propónse a extensión dunha librería de checkpointing para facilitares a implementación de estratexias de recuperación ad hoc ante conupcións de memoria.
En moitas ocasións, estos erros poden ser xestionados a nivel de aplicación, evitando
desencadear un fallo-parada e permitindo unha recuperación máis eficiente.[Resumen]
El rápido aumento de las necesidades de cómputo de distintas ramas de la ciencia
ha provocado un gran crecimiento en el rendimiento ofrecido por los supercomputadores.
Cada vez se construyen sistemas de computación de altas prestaciones mayores,
con más recursos hardware de distintos tipos, lo que hace que las tasas de
fallo del sistema aumenten. Por tanto, el estudio de técnicas de tolerancia a fallos
eficientes resulta indispensable para garantizar que los programas científicos puedan
completar su ejecución, evitando además que se dispare el consumo de energía. La
técnica checkpoint/restart es una de las más populares. Sin embargo, la mayor parte
de la investigación en este campo se ha centrado en estrategias stop-and-restart
para aplicaciones de memoria distribuida tras la ocurrencia de fallos-parada. Esta
tesis propone técnicas checkpoint/restart a nivel de aplicación para los modelos de
programación paralela más populares en supercomputación. Se han implementado
protocolos de checkpointing para aplicaciones híbridas MPI-OpenMP y aplicaciones
heterogéneas basadas en OpenCL, prestando en ambos casos especial atención a la
portabilidad y la maleabilidad de la solución. Con respecto a aplicaciones de memoria
distribuida, se propone una solución de resiliencia que puede ser usada de forma
genérica en aplicaciones MPI SPMD, permitiendo detectar y reaccionar a fallosparada
sin abortar la ejecución. En su lugar, se vuelven a lanzar los procesos fallidos
y se recupera el estado de la aplicación con una vuelta atrás global. A mayores, esta
solución de resiliencia ha sido optimizada implementando una vuelta atrás local, en
la que solo los procesos fallidos vuelven atrás, empleando un protocolo de almacenaje
de mensajes para garantizar la consistencia y el progreso de la ejecución. Por
último, se propone una extensión de una librería de checkpointing para facilitar la
implementación de estrategias de recuperación ad hoc ante corrupciones de memoria.
Muchas veces, este tipo de errores puede gestionarse a nivel de aplicación, evitando
desencadenar un fallo-parada y permitiendo una recuperación más eficiente.[Abstract]
The rapid increase in the computational demands of science has lead to a pronounced
growth in the performance offered by supercomputers. As High Performance
Computing (HPC) systems grow larger, including more hardware components
of different types, the system's failure rate becomes higher. Efficient fault
tolerance techniques are essential not only to ensure the execution completion but
also to save energy. Checkpoint/restart is one of the most popular fault tolerance
techniques. However, most of the research in this field is focused on stop-and-restart
strategies for distributed-memory applications in the event of fail-stop failures. Thís
thesis focuses on the implementation of application-level checkpoint/restart solutions
for the most popular parallel programming models used in HPC. Hence, we
have implemented checkpointing solutions to cope with fail-stop failures in hybrid
MPI-OpenMP applications and OpenCL-based programs. Both strategies maximize
the restart portability and malleability, ie., the recovery can take place on
machines with different CPU / accelerator architectures, and/ or operating systems,
and can be adapted to the available resources (number of cores/accelerators). Regarding
distributed-memory applications, we propose a resilience solution that can
be generally applied to SPMD MPI programs. Resilient applications can detect and
react to failures without aborting their execution upon fail-stop failures. Instead,
failed processes are re-spawned, and the application state is recovered through a
global rollback. Moreover, we have optimized this resilience proposal by implementing
a local rollback protocol, in which only failed processes rollback to a previous
state, while message logging enables global consistency and further progress of the
computation. Finally, we have extended a checkpointing library to facilitate the
implementation of ad hoc recovery strategies in the event of soft errors) caused by
memory corruptions. Many times, these errors can be handled at the software-Ievel,
tIms, avoiding fail-stop failures and enabling a more efficient recovery
- …
