Tareador: The Unbearable Lightness of Exploring Parallelism

Subotic, Vladimir; Campos, Arturo Salazar; Velasco, Alejandro; Ayguadé, Eduard; Labarta, Jesús; Valero, Mateo

doi:10.1007/978-3-319-16012-2_4

Cited by 4 publications

(2 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The graphs in Figure 3 and Figure 4, show how the emergency service is parallelized for three different scenarios: i) cloud (i.e., the traditional cloud scenario excluding fog devices, shown to depict the F2C benefits); ii) fog-to-cloud (F2C), and; iii) optimized fog-to-cloud (OF2C). The parallelization has been obtained applying the Tareador tool [9], developed at the Barcelona Supercomputing Center (BSC). The tool is widely used in the High Performance Computing field to assist parallelization of sequential applications by decomposing sequential specification into tasks and dependencies facilitating parallel execution.…”

Section: IIImentioning

confidence: 99%

Foggy clouds and cloudy fogs: a real need for coordinated management of fog-to-cloud computing systems

Masip-Bruin

Marín‐Tordera

Tashakor

et al. 2016

IEEE Wireless Commun.

255

114

View full text Add to dashboard Cite

-The recent advances in the cloud services technology are fueling a plethora of information technology innovation, including networking, storage and computing. Today, various flavors have evolved of Internet of Things (IoT), cloud computing and the so-called fog computing, -a concept referred to capabilities of edge-devices and user's clients to compute, store and exchange data among each other and with the cloud. Though the evolution was not easily foreseeable to happen at such a rapid pace, each piece of it today facilitates and enables the deployment of what we commonly refer to as a smart scenario, including smart cities, smart transportation and smart homes. As most of the cloud, fog and network services today run simultaneously in each scenario, we observe that we are at the dawn of what maybe the next big step in the cloud computing and networking evolution, whereby services might be executed at the network edge, both in parallel and in a coordinated fashion, as well as supported by the unstoppable technology evolution. As edge devices become richer in functionality and smarter, -embedding capacities such as storage or processing, as well as embedding new functionalities, such as decision making, data collection and forwarding, sharing, etc, a real need is emerging for coordinated management of fog-to-cloud (F2C) computing systems. This paper introduces a layered fog-to-cloud (F2C) architecture, its benefits and strengths as well as the arising open and research challenges, making the case for the real need for their coordinated management. Our architecture, the illustrative use case presented and a comparative performance analysis, albeit conceptual, all clearly show the way forward towards a new IoT scenario with a set of existing and unforeseen services provided on a highly distributed and dynamic compute, storage and networking resources, bringing together heterogeneous and commodity edge devices, emerging fogs as well as conventional clouds. Keywords-Cloud computing, fog computing, fog-to-cloud, Internet of Things (IoT) I. INTRODUCTION: THE SCENARIOThe most recent developments in the information and communications technologies area have started to make a profound impact, through massive connectivity of humans and computers, as well as a massive proliferation of edge devices carried by humans (i.e., smart phones, and those associated with all the surroundings -the Internet of Things). These two major commodities not only have facilitated the true "anywhere, anyhow, anytime" users' connectivity, but also the data collection, further enabling the deployment of new value-added services. Today, the scenarios of smart cities, smart transportation and smart homes are no more domain of research of distant future, but are becoming the new "normal". Several references can be found in the literature that already showed the notable effect these concepts can bring to the business market [1]. For a rapid business and technological success to happen, however, two inherent features need to be addressed in t...

show abstract

Section: IIImentioning

confidence: 99%

Foggy clouds and cloudy fogs: a real need for coordinated management of fog-to-cloud computing systems

Masip-Bruin

Marín‐Tordera

Tashakor

et al. 2016

IEEE Wireless Commun.

255

114

View full text Add to dashboard Cite

show abstract

“…The proposed programming model has been used in many research projects by users and showed potential to replace already accepted programming approaches. The work of thesis became a motivation for tool developers to provide better support for porting [46] and debugging [8] hybrid MPI/OmpSs programming model.…”

Section: Discussionmentioning

confidence: 99%

The MPI/OmpSs parallel programming model

Marjanović

View full text Add to dashboard Cite

Even today supercomputing systems have already reached millions of cores for a single machine, which are connected by using a complex network interconnection. Reducing communication time across processes becomes the most important issue in order to achieve the highest possible performance. The Message Passing Interface (MPI), which is the most widely used programming model for large distributed memory, supports asynchronous communication primitives for overlapping communication and computation. However, these primitives are difficult to use and increase code complexity. which then requiring more development effort and making less readable programs. This thesis presents a new programming model, which allows the programmer to easily introduce the asynchrony necessary to overlap communication and computation. The proposed programming model is based on MPI and tasked based shared memory framework, namely OmpSs. The thesis further describes implementation details which in order to allow efficient inter-operation of the OmpSs runtime and MPI. The thesis demonstrates the hybrid use of MPI/OmpSs with several applications of which the HPL benchmark is the most important case study. The hybrid MPI/OmpSs versions significantly improve the performance of the applications compared with their pure MPI counterparts. For the HPL we get close to the asymptotic performance at relatively small problem sizes and still get significant benefits at large problem sizes. In addition, the hybrid MPI/OmpSs approach substantially reduces code complexity and is less sensitive to network bandwidth and operating system noise than the pure MPI versions. In addition, the thesis analyzes and compares current techniques for overlapping computation and collective communication, including approaches using point-to-point communications and additional communication threads, respectively. The thesis stresses the importance of understanding the characteristic of a computational kernel that runs concurrently with communication. Experimental evaluations is done using the Communication Computation Concurrent (CCUBE) synthetic benchmark, developed in this thesis, as well as the HPL. Las supercomputadoras están formadas por un creciente número de núcleos, del orden de millones en la actualidad, que se comunican a través de una compleja red de interconexión. Para obtener el más alto rendimiento posible es necesario reducir el tiempo de comunicación entre procesos. MPI ("Message Passing Interface", Interfaz de Paso de Mensajes), el modelo de programación más usado para grandes sistemas con memoria distribuida, permite llamadas de comunicación asíncrona para solapar la comunicación y la computación. Sin embargo, dichas llamadas son difíciles de usar e incrementan la complejidad del código, necesitándose un mayor esfuerzo en la implementación del código y dando lugar a programas más difíciles de leer. Esta tesis presenta un nuevo modelo de programación que permite al programador introducir fácilmente la asincronía necesaria para solapar la comunicación y la computación. El modelo de programación propuesto está fundamentado en MPI y la infraestructura basada en tareas y memoria compartida OmpSs. La tesis describe en profundidad los detalles de la implementación para la eficiente interoperabilidad entre OmpSs y MPI. En la tesis se demuestra el uso híbrido de MPI/OmpSs con distintas aplicaciones de las cuales el benchmark HPL es el más importante. La versión híbrida MPI/OmpSs mejora significativamente el rendimiento de las aplicaciones respecto a las versiones MPI originales. En el caso de HPL se acerca a un rendimiento asintótico para problemas relativamente pequeños, obteniendo mejoras significativas para problemas grandes. Además la versión híbrida MPI/OmpSs reduce substancialmente la complejidad del código y se ve menos afectada por el ancho de banda de la red y el ruido del sistema operativo que la versión MPI pura. Esta tesis también analiza y compara otros métodos actuales para solapar computación y comunicación colectiva, tales como usar comunicación punto a punto con hilos adicionales para la comunicación. La tesis resalta la importancia de entender las características de la computación que se ejecuta simultáneamente con la comunicación. Los resultados experimentales se han obtenido usando el benchmark sintético CCUBE ("Communication Computation Concurrent", Comunicación Computación Concurrente), desarrollado en esta tesis, además de HPL.

show abstract

Addressing HPC skills shortages with parallel computing MOOC

Sancho

Alexandrova²,

González³

2015

2015 International Conference on Interactive Collaborative and Blended Learning (ICBL)

View full text Add to dashboard Cite

The technical inflection points in the route to Exascale and the existing talent gap in Computational Science and HPC are well publicized. In this paper is described the development of a xMOOC format course with the aim at reaching a target audience composed of BSc, MSc and professional training students that have no experience in parallel programming, as one approach to deal with the lack of specialists that can tackle parallelism. This MOOC utilizes UCATx platform based on and linked to edX. Relying on the expertise of BSC Education and Training Team and UPC Computer Science Lecturers, was devised a course syllabus at introductory level. The paper outlines the design of the course and highlights some of the problems to deliver on-line courses on technical subjects while highlighting the developers' approach to tackle them. An evaluation process for the course is described and the current trends of MOOC courses are reviewed.Peer ReviewedPostprint (published version

show abstract

Tareador: The Unbearable Lightness of Exploring Parallelism

Cited by 4 publications

References 14 publications

Foggy clouds and cloudy fogs: a real need for coordinated management of fog-to-cloud computing systems

Foggy clouds and cloudy fogs: a real need for coordinated management of fog-to-cloud computing systems

The MPI/OmpSs parallel programming model

Addressing HPC skills shortages with parallel computing MOOC

Contact Info

Product

Resources

About