De Giusti, Armando Eduardo
The purpose of this book is to break down proven and mature cloud computing technologies and practices into a series of well-defined concepts, models, architectures and technology mechanisms.
Extracting the performance characteristics of the diﬀerent I/O conﬁgurations of the I/O system.
These activities are independent. The characterization of application is done oﬀ-line and the application I/O model can be applied to analyze diﬀerent target systems. The application I/O model is deﬁned by three characteristics: metadata, spatial global pattern and temporal global pattern. The I/O model of the application is expressed by I/O phases, where an I/O phase is a repetitive sequence of same pattern on a ﬁle for a number of processes of the parallel application. A phase will be signiﬁcant depending on data transferred that represent its weight.
Costa, André Luiz Lima da; Souza, Josemar Rodrigues de
Given the availability of computer clusters based on multi-core processors, the hybrid programming model has become an important ally of high-performance computing users in improving the performance of their parallel applications. However, creating hybrid applications is a complex task because it requires developers to be familiar with two distinct parallel programming models. Against this background, this article introduces APCM, an auto-parallelism computational model. APCM’s goal is to create hybrid parallel applications, i.e., OpenMP (memory programming) and a message-passing interface (MPI), from MPI applications. This goal is achieved in a simple, automated manner that is transparent for the user while increasing...
García Garino, Carlos; Pacini, Elina; Monge, David A.; Careglio, Claudio; Mirasso, Aníbal
Cloud computing promises great opportunities for the execution of scienti c and engineering applications.
However, the execution of such kind of applications over Cloud infrastructures requires the accomplishment of many complex processes. In this paper we present a Computational Mechanics Software as a Service (SaaS) project which will allow scientists to easily con gure and submit their experiments to be transparently executed on the Cloud. For this purpose, a nite element software called SOGDE is used to perform parametric studies of computational mechanics on the basis of underlying computing resources.
Moreover, a web service provides an interface for the abovementioned functionalities allowing...
Expósito, Roberto R.; Taboada, Guillermo L.; Pardo, Xoán C.; Touriño, Juan; Doallo, Ramón
Amazon Web Services (AWS) is a well-known public Infrastructure-as-a-Service (IaaS) provider whose Elastic Computing Cloud (EC2) o ering includes some instances, known as cluster instances, aimed at High-Performance Computing (HPC) applications. In previous work, authors have shown that the scalability of HPC communication-intensive applications does not bene t from using higher computational power cluster instances as much as it could be expected.
Cost analysis recommends using lower computational power cluster instances unless high memory requirements preclude their use. Moreover, it has been observed that scalability is very poor when more than one instance is used due to network virtualization overhead. Based...
Murazzo, María A.; Rodríguez, Nelson R.; Villafañe, Daniela A.; González, Facundo
Nowadays it is easy for the users who live in a digital era, to access social networks not only from their computers, but also from their mobile phones, to upload pictures instantaneously, to send messages on Whatsapp or to share their location with other users.
At the same time, the business world is also changing at highspeed.
Dematerialization has already affected different business sectors; it has even put an end to a model that has been used for years.
Cloud Computing allows technology to access in the economy and society, not only allowing users to be connected to this new digital world through...
Rodríguez, Nelson R.; Murazzo, María A.; Chávez, Susana Beatriz; Valenzuela, Francisca Adriana; Martín, Adriana Elba; Villafañe, Daniela A.
Mobile networking has changed people's lifestyles and the business world drastically. It can be widely implemented and with the ability to use this mobile computing, companies can increase their productivity greatly through this practice. Devices and mobile networks have limitations, such as battery life or processing speed, these further hinder the utilization of this method of work. Cloud Computing is an application that reduces the impact of the limitations that mobile computing presents. However, besides the characteristics of mobile computing as well as the technological advances, such as mobile TV or workplaces that implement BYOD, clearly differentiate even more Cloud...
Santos, Juan E.; Gauzellino, Patricia M.; Savioli, Gabriela B.; Martínez Corredor, Robiel
This paper presents a collection of finite element procedures to model seismic wave propagation at the macroscale taking into account the effects caused by heterogeneities occuring at the mesoscale. For this purpose we first apply a set of compressibility and shear experiments to representative samples of the heterogeneous fluid saturated material. In turn these experiments yield the effective coefficients of an anisotropic macroscopic medium employed for numerical simulations at the macroscale. Numerical experiments illustrate the implementation of the proposed methodology to model wave propagation at the macroscale in a patchy brine-CO2 saturated porous medium containing a dense set of parallel...
Balladini, Javier; Muresano, Ronal; Suppi, Remo; Rexachs del Rosario, Dolores; Luque Fadón, Emilio
Over the last decade, the computing clusters have been updated in order to satisfy the increasing demand of greater computational power for running applications.
However, this increasing is transformed in more system en- ergy consumption, which results in financial, environmental and in some cases with social consequences. Hence, the ideal is to achieve an scenario that allows the system admin- istrator to find a trade-off between time and energy-efficiency for parallel algorithms on virtualized environments. The main objective of this work is based on developing an analytical model to predict the energy consumption and energy delay product (EDP) for SPMD applications...
Muresano, Ronal; Wong, Alvaro; Rexachs del Rosario, Dolores; Luque Fadón, Emilio
Progress in the parallel programming field has allowed scientific applications to be developed with more complexity and accuracy. However, such precision requires greater computational power in order to be executed. How- ever, updating the local systems could be considered an expensive decision. For this reason, cloud computing is emerging as a commercial infrastructure that allows us to eliminate maintaining the computing hardware. For this reason, cloud is promising to be a computing alternative to clusters, grids and supercomputing for executing these applications. In this sense, this work is focused on describing the manner of migrating our prediction tool PAS2P (parallel...
Rodríguez, Christian Adrián; Molinari, Lía; Díaz, Francisco Javier
The coexistence of multiple platforms and the implementation of different virtualization models makes server administration more complex every day. The undeniable benefits both methodologies offer in terms of performance optimization and energy saving can be overshadowed if clear guidelines are not established for configuration and maintenance in accordance with the needs of increasingly agile development models that demand quick responses. DevOps is a possible solution to this situation. However, it demands a new perspective in traditional roles within technology areas.
Barrionuevo, Mercedes; Britos, Luís; Bustos, Fabricio H.; Gil Costa, Graciela Verónica; Lopresti, Mariela; Mancini, Virginia; Miranda, Natalia; Ochoa, César; Piccoli, María Fabiana; Printista, Alicia Marcela; Reyes, Nora Susana
With the technology advance and the growth of Internet, the information that can be found in this net, as well as the number of users that access to look for speciﬁc data is bigger. Therefore, it is desirable to have a search system that allows to retrieve information at a reasonable time and in an efﬁcient way. In this paper we show two computing paradigms appropriate to apply in the treatment of large amounts of data consisting of objects such as images, text, sound and video, using hybrid computing over MPI+OpenMP and GPGPU.
The proposal is developed through experience gained in...