Filters








8,516 Hits in 6.0 sec

Embedded Systems with Component-Based GPU Support: A State of the Art

A. Miclăuș, Department of Computer Science, Faculty of Mathematics and Computer Science, Babeș-Bolyai University, Cluj-Napoca, Romania, Ș. Petrescu, A. Vescan, Department of Computer Science, Faculty of Mathematics and Computer Science, Babeș-Bolyai University, Cluj-Napoca, Romania, Department of Computer Science, Faculty of Mathematics and Computer Science, Babeș-Bolyai University, Cluj-Napoca, Romania
2019 Studia Universitatis Babes-Bolyai: Series Informatica  
However, developing embedded systems with GPUs by using CBD was considered until very recently a problem with restricted availability and flexibility.  ...  Two trends are present in the field: the usage of boards with Graphics Processing Units (GPUs) and the usage of component-based development (CBD).  ...  function of optimization criteria [4] , [11] CONCLUSIONS Facilitation of component-based development of embedded systems with GPUs is a need in alternative finding of solutions for high-demand  ... 
doi:10.24193/subbi.2019.1.06 fatcat:nlhabzyz2jaajmfchurpowa5tq

Component Allocation Optimization for Heterogeneous CPU-GPU Embedded Systems

Gabriel Campeanu, Jan Carlson, Severine Sentilles
2014 2014 40th EUROMICRO Conference on Software Engineering and Advanced Applications  
In this paper, we present a formal model for allocation optimization of embedded systems which contains a mix of CPU and GPU processing nodes.  ...  In its current version, optimized allocation schemes are generated through an integer programming technique to balance the system resource utilization and to optimize the system performance using the GPU  ...  ACKNOWLEDGMENT Our research is supported by the RALF3 project (http://www.mrtc.mdh.se/projects/ralf3/) through Swedish Foundation for Strategic Research (SSF).  ... 
doi:10.1109/seaa.2014.29 dblp:conf/euromicro/CampeanuCS14 fatcat:gmc65jtitzd3xjkz7lrcsfj24e

Optimized Realization of Software Components with Flexible OpenCL Functionality

Gabriel Campeanu, Jan Carlson, Séverine Sentilles
2018 Proceedings of the 13th International Conference on Evaluation of Novel Approaches to Software Engineering  
Today, the newly available embedded boards with GPUs provide a solution to satisfy the ever-increasing requirements of modern embedded systems.  ...  Component-based development is a well-known paradigm used to develop embedded systems. However, this paradigm lacks GPU support to address the specifics of these new boards.  ...  ACKNOWLEDGMENTS The Swedish Foundation for Strategic Research (SSF) supports our work through the RALF3 project (IIS11-0060).  ... 
doi:10.5220/0006691500770088 dblp:conf/enase/CampeanuCS18 fatcat:brdmf6zjevbejeb7vx6coyskui

Support for high performance using heterogeneous embedded systems

Gabriel Campeanu
2013 Proceedings of the 18th international doctoral symposium on Components and architecture - WCOP '13  
Nowadays it is more common to build embedded system on a heterogeneous platform, i.e. a platform containing different computational units such as mCPU, GPU and FPGA.  ...  For different constraints and requirements, different deployment configuration can be optimal.  ...  RELATED WORK In the first part of this section, we present an overview of different component-based and model-driven models for embedded systems and how they manage to provide support for EFPs.  ... 
doi:10.1145/2465498.2465502 fatcat:l4yaefv7hvesxghieiauq3njcq

Parallel Execution Optimization of GPU-aware Components in Embedded Systems

Gabriel Campeanu
2017 Proceedings of the 29th International Conference on Software Engineering and Knowledge Engineering  
In the context of component-based CPU-GPU embedded systems, we propose an automatic method that provides parallel execution schemes of components with GPU capabilities.  ...  Moreover, the method optimizes the overall system performance based on component execution times and system architecture (i.e., communication pattern).  ...  ACKNOWLEDGMENTS The Swedish Foundation for Strategic Research (SSF) supports our work inside the RALF3 project (IIS11-0060).  ... 
doi:10.18293/seke2017-137 dblp:conf/seke/Campeanu17 fatcat:36s4r7eg4jacjexsm43y2my7mi

Towards a Complete Model for Software Component Deployment on Heterogeneous Platform

Ivan Švogor
2014 Research Papers. Faculty of Materials Science and Technology. Slovak University of Technology in Trnava  
This report briefly describes an ongoing research related to optimization of allocating software components to heterogeneous computing platform (which includes CPU, GPU and FPGA).  ...  of the problem at hand and help developers determine component placement in the system design phase.  ...  Current insights We proposed an extension of our previous model (12) for optimization of component allocation on a heterogeneous embedded platform.  ... 
doi:10.2478/rput-2014-0005 fatcat:agz5hrrvunf6vgxvnklr33lvxq

Optimized Deep Neural Networks for Real-Time Object Classification on Embedded GPUs

Syed Rizvi, Gianpiero Cabodi, Gianluca Francini
2017 Applied Sciences  
In this paper, a matrix multiplication-based convolution (ConvMM) approach is fully parallelized using concurrent resources of GPU (Graphics Processing Unit) and optimized, considerably improving the performance  ...  The performance of this optimized and accelerated convolutional layer is compared with its sequential and heterogeneous versions.  ...  Proposed Unified Memory-based Allocations for GPU-Only ConvMM and Other Layers Typical desktop systems have the separate host (CPU) and device (GPU) memories as shown in Figure 6 .  ... 
doi:10.3390/app7080826 fatcat:i7fwrv7umjcorhl7hwm3qnyrby

An Improved Coloured Petri Net Model for Software Component Allocation on Heterogeneous Embedded Systems

Issam Al-Azzoni
2018 Journal of Computing and Information Technology  
We improve the CPN model for the embedded systems and outline a technique that exploits CPN Tools, a well-known CPN tool, to efficiently analyze embedded system's state space and find optimal allocations  ...  We extend an approach to component allocation on heterogeneous embedded systems using Coloured Petri Nets (CPNs).  ...  The authors of [1] present a formal model for allocation optimization of embedded systems which contain a mix of CPU and GPU processing nodes.  ... 
doi:10.20532/cit.2018.1003892 fatcat:hpyrrhsilbbhfemwqf3vktfwdq

A Modeling Approach based on UML/MARTE for GPU Architecture [article]

Antonio Wendell De Oliveira Rodrigues , Frédéric Guyomarc'H, Jean-Luc Dekeyser (INRIA Lille - Nord Europe)
2011 arXiv   pre-print
Nowadays, the High Performance Computing is part of the context of embedded systems.  ...  The results show that this approach will help to generate code for GPUs based on model transformations using Model Driven Engineering (MDE).  ...  Thus, in order to optimize their application performance, embedded systems can use the processing power of GPUs.  ... 
arXiv:1105.4424v1 fatcat:e2csase72bh2vlrldzwkpam3iu

Meta-Heuristics for Solving the Software Component Allocation Problem

Issam Al-Azzoni, Saqib Iqbal
2020 IEEE Access  
INDEX TERMS Component allocation, model-driven engineering, embedded systems, heterogeneous systems, genetic algorithms, ant colony optimization.  ...  The software component allocation problem is concerned with mapping a set of software components to the computational units available in a heterogeneous computing system while maximizing a certain objective  ...  [29] , [30] provides an optimization model for allocating components in heterogeneous CPU-GPU Embedded Systems.  ... 
doi:10.1109/access.2020.3015864 fatcat:rdaorgq6undexoxzkcrnhpwr5q

An initial performance review of software components for a heterogeneous computing platform

Ivan Švogor
2015 Proceedings of the 2015 European Conference on Software Architecture Workshops - ECSAW '15  
The design of embedded systems is a complex activity that involves a lot of decisions.  ...  This paper presents power consumption and execution time profiling of a component software that can be allocated on heterogeneous computing units (CPU, GPU, FPGA) of a tracked robot.  ...  INTRODUCTION The design of present day embedded systems and the accompanying software is becoming ever more complex, and with an emergence of specialized computing units for accelerating certain operations  ... 
doi:10.1145/2797433.2797449 dblp:conf/ecsa/Svogor15 fatcat:o4heiipmjjenvdflq7sx4763ja

Automatic Synthesis of Heterogeneous CPU-GPU Embedded Applications from a UML Profile

Federico Ciccozzi
2013 ACM/IEEE International Conference on Model Driven Engineering Languages and Systems  
Modern embedded systems present an ever increasing complexity and model-driven engineering has been shown to be helpful in mitigating it.  ...  CPU-GPU configurations.  ...  In [1] we proposed an automated round-trip engineering support for MDE of embedded systems with focus on the preservation of extra-functional properties (EFPs).  ... 
dblp:conf/models/Ciccozzi13 fatcat:epustqhez5fanasiyx2v6tmiae

Martin-CSE21.pdf [article]

Daniel Martin
2021 figshare.com  
While they are linked by the basic nature of their problems -- all are fundamentally based on solving systems of partial differential equations (PDEs), the applications have adopted a wide range of approaches  ...  Not visible to the application. o Replaces Fortran call-backs on CPU-based systems with 10x reduction in code size. o High performance is obtained for data resident on GPUs. • Data on unions of rectangles  ...  , sum, etc.) • Interacts with Fortran -Wrap existing Fortran allocations in YAKL Fortran-like Arrays -Pool allocator with CUDA Managed Memory hooks and Fortran hooks -Enables an incremental porting path  ... 
doi:10.6084/m9.figshare.14153717.v1 fatcat:kioz7cyzmrdntbriwp7akp6une

Resource Management for Improving Overall Reliability of Multi-Processor Systems-on-Chip [chapter]

Yue Ma, Junlong Zhou, Thidapat Chantem, Robert P. Dick, X. Sharon Hu
2020 Embedded Systems  
AbstractMulti-processor systems on a chip (MPSoCs) are widely deployed in real-time embedded systems.  ...  In this chapter, we first introduce concepts and models associated with the two reliability metrics, then present two techniques that optimize them separately.  ...  [13] adopt mean time to failure (MTTF) as the common metric to evaluate SER and LTR and design a success tree based scheme for reliability analysis for embedded systems.  ... 
doi:10.1007/978-3-030-52017-5_10 fatcat:arcgfac4g5crfl6l26nvszj73m

Co-Optimizing Performance and Memory FootprintVia Integrated CPU/GPU Memory Management, anImplementation on Autonomous Driving Platform [article]

Soroush Bateni, Zhendong Wang, Yuankun Zhu, Yang Hu, Cong Liu
2020 arXiv   pre-print
Cutting-edge embedded system applications, such as self-driving cars and unmanned drone software, are reliant on integrated CPU/GPU platforms for their DNNs-driven workload, such as perception and other  ...  Based on this observation, we develop a performance model that can predict system overhead for each memory management method based on application characteristics.  ...  embedded systems provides a reasonable trade-off between performance, energy usage, thermal capacity, and space usage compared to a discrete GPU component [1] .  ... 
arXiv:2003.07945v2 fatcat:rqfnjujirfgz7eftz5q3uyoh6y
« Previous Showing results 1 — 15 out of 8,516 results