This paper presents the architecture of the I/O subsystem for mass storage which has been proposed for MULTIPLUS, a high performance scientific multiprocessor under development at NCE/UFRJ. After studying the I/O bottleneck problem, which results from an existing gap in the technological evolution of processor and mass storage devices, and identifying the parameters which define the I/O subsystem performance, the paper discusses the advantages and drawbacks of the use of a concentrated or distributed I/O subsystem architecture. Finally, a distributed architecture, which is suitable for scientific parallel application, is proposed for the MULTIPLUS I/O subsystem and the internal organization of the I/O processosr is presented and justified t...
In order to achieve performance gains in the software, computers have evolvedto multi-core and many-...
Grid computing enables organizations to integrate and share sets of heterogeneous resources into on...
In this thesis we propose a new simulation platform specifically designed for modeling parallel and ...
This paper discusses some issues concerning the implementation of Lee's routing algorithm on MULTIPL...
Increasing processors' clock frequency has traditionally been one of the largest drivers of performa...
This article presents a definition of the basic model for parallel programming that will be provided...
Exploiting concurrency to achieve greater performance is a difficult and important challenge for cur...
The MULTIPLUS project is currently under development at NCE/UFRJ, Brazil, aims at the study of paral...
Recent advances in storage technologies and high performance interconnects have made possible in the...
The application constraints driving the design of embedded systems are constantly demanding higher p...
The computational requirements of the processors are increasing tremendously with the increase in th...
Since several years, classical multiprocessor systems have evolved to multicores, which tightly inte...
Today, the processing of large dataset is generally parallelised and performed on computers with man...
The current parallel architectures integrate processors with many cores to shared memory growing and...
Dissertação de mestrado em Engenharia InformáticaMost event data analysis tasks in the ATLAS project...
In order to achieve performance gains in the software, computers have evolvedto multi-core and many-...
Grid computing enables organizations to integrate and share sets of heterogeneous resources into on...
In this thesis we propose a new simulation platform specifically designed for modeling parallel and ...
This paper discusses some issues concerning the implementation of Lee's routing algorithm on MULTIPL...
Increasing processors' clock frequency has traditionally been one of the largest drivers of performa...
This article presents a definition of the basic model for parallel programming that will be provided...
Exploiting concurrency to achieve greater performance is a difficult and important challenge for cur...
The MULTIPLUS project is currently under development at NCE/UFRJ, Brazil, aims at the study of paral...
Recent advances in storage technologies and high performance interconnects have made possible in the...
The application constraints driving the design of embedded systems are constantly demanding higher p...
The computational requirements of the processors are increasing tremendously with the increase in th...
Since several years, classical multiprocessor systems have evolved to multicores, which tightly inte...
Today, the processing of large dataset is generally parallelised and performed on computers with man...
The current parallel architectures integrate processors with many cores to shared memory growing and...
Dissertação de mestrado em Engenharia InformáticaMost event data analysis tasks in the ATLAS project...
In order to achieve performance gains in the software, computers have evolvedto multi-core and many-...
Grid computing enables organizations to integrate and share sets of heterogeneous resources into on...
In this thesis we propose a new simulation platform specifically designed for modeling parallel and ...