In this paper a comparative study between One-Class-One-Network (OCON) and Multi-Layered Perceptron (MLP) neural networks for vowel phoneme recognition is presented. the OCON architecture, first proposed by *** et al ...
详细信息
In this paper a prototype of an optical multi-faceted Free-Space Image Distributor for optical interconnection in massively parallelprocessing is presented. the consecutive aspects of the design of a free space orien...
详细信息
ISBN:
(纸本)0818685727
In this paper a prototype of an optical multi-faceted Free-Space Image Distributor for optical interconnection in massively parallelprocessing is presented. the consecutive aspects of the design of a free space oriented image distributor are considered. Features of free space image distributing techniques are studied and solutions are proposed that favour the development, prototyping and testing of a free space oriented 9-faceted image distributing system (the Kaleidoscope) for massively parallelprocessing. Optically modelling the kaleidoscopic system and simulating the optical backplane are carried out. Multiple complementary alternatives in designing a multi- faceted (N>9) kaleidoscopic system are analysed.
the paper focuses on local load balancing policies far massively parallelarchitectures and introduces a new scheme for load information exchange between neighbor nodes. the idea is to distort the exchanged load infor...
详细信息
ISBN:
(纸本)0818691948
the paper focuses on local load balancing policies far massively parallelarchitectures and introduces a new scheme for load information exchange between neighbor nodes. the idea is to distort the exchanged load information to let the policy keep into account a more global view of the system and overcome the limits of the local scope. the presented scheme has been integrated into two variants of a direct-neighbor policy and evaluated in dependence of the characteristics of the system load. Experimental results show that the transmission of distorted load information provides high efficiency unless the dynamicity of the load becomes too high, in which case it is preferable to exploit nondistorted toad information.
We analyzed the performance improvement of parallelprocessing systems with fully or highly connected optical interconnections over systems with a bus. Solving an 1024 by 1024 point 2-D FFT with 300Mflops processors a...
详细信息
ISBN:
(纸本)0818685727
We analyzed the performance improvement of parallelprocessing systems with fully or highly connected optical interconnections over systems with a bus. Solving an 1024 by 1024 point 2-D FFT with 300Mflops processors and 66MHz interconnections, the speedup factor of the execution time of fully interconnected systems over systems with a bus is 3.1 for 32 PEs, and 1.2 for 4 PEs. We proposed an extended ring system as a highly connected interconnection, and also discussed the speed-up factor as for highly connected interconnections.
Presented in this paper is an implementation using a combination of successive pruning algorithm[I] and a parallel architecture using a digital signal processor and a general purpose processor Ibr progressive transmis...
详细信息
ISBN:
(纸本)0818691948
Presented in this paper is an implementation using a combination of successive pruning algorithm[I] and a parallel architecture using a digital signal processor and a general purpose processor Ibr progressive transmission of still images. the adaptive pruning algorithm[2] is used for ensuring a minimum quality of the image while further progressions on the image are computed using a modified successive pruning algorithm. Due to the reduction in computations achieved using successive pruning and efficient splitting of tasks between the two processors, the speed of the entire operation is found to increase by a factor of 2.8 compared to the normal implementation of the progressive mode of JPEG.
the proceedings contain 28 papers. the topics discussed include: high-performance electrical signaling;entrepeneurship is a career option;integrated free-space optical interconnects for chip-to-chip communications;opt...
ISBN:
(纸本)0818685727
the proceedings contain 28 papers. the topics discussed include: high-performance electrical signaling;entrepeneurship is a career option;integrated free-space optical interconnects for chip-to-chip communications;optical multi-faceted free-space image distributor for massively parallelprocessing;translucent smart pixel array (TRANSPAR) chips for high throughput networks and SIMD signal processing;a free space optical crossbar switch using wavelength division multiplexing and vertical-cavity surface emitting lasers;strategies for competitive optical interconnection technologies;opportunities for optical communications in petaflops-scale computing;multicast communication in a class of rearrangeable optical WDM network;optimal routing and channel assignments for hypercube communication on optical mesh-like processor arrays;multicast control in optical circuit-switched banyan networks;performance improvement of parallelprocessing systems with fully or highly connected optical interconnections over systems with a bus;channel allocation, power budget and bit error rate in hierarchical optical ring interconnection network (HORN);a new analysis for wavelength translation in regular WDM networks;and a high-performance router architecture for multimedia applications.
this paper proposes a modified parameter mapping scheme for parallel model combination (PMC) method. the modification aims to improve the discriminative capabilities of the compensated models. It is achieved by the re...
详细信息
Several recent architectural trends appear to require some of the special properties of optical interconnection networks. the memory hierarchy will probably get deeper and the hardware will be more dynamic. Reconfigur...
详细信息
ISBN:
(纸本)0818685727
Several recent architectural trends appear to require some of the special properties of optical interconnection networks. the memory hierarchy will probably get deeper and the hardware will be more dynamic. Reconfigurable hardware and software are the subject of intense research. It is likely that the basic word size and the number of communication streams will be dynamically configurable. Optical interconnection networks, and in particular, bit parallel free-space implementations, may provide the necessary flexibility.
High Performance Fortran (HPF) is the de facto standard language for writing data parallel programs. In case of applications that use indirect addressing on distributed arrays, HPF compilers have limited capabilities ...
详细信息
ISBN:
(纸本)3540649522
High Performance Fortran (HPF) is the de facto standard language for writing data parallel programs. In case of applications that use indirect addressing on distributed arrays, HPF compilers have limited capabilities for optimizing such codes on distributed memory architectures, especially for optimizing communication and reusing communication schedules between subroutine boundaries. this paper describes a dynamic approach for optimizing unstructured communication in codes with indirect addressing. the basic idea is that runtime data reflecting the communication patterns will be reused if possible. the user has only to specify which data in the program has to be traced for modifications. the experiments and results show the effectiveness of the chosen approach.
this paper discusses the main achievements of the EPIC project, whose aim was to design a high level programming environment with an associated implementation for portable parallel image processing. the project was fu...
详细信息
ISBN:
(纸本)3540649522
this paper discusses the main achievements of the EPIC project, whose aim was to design a high level programming environment with an associated implementation for portable parallel image processing. the project was funded as part of the EPSRC Portable Software Tools for parallelarchitectures (PSTPA) programme. the paper summarises new portable programming abstractions for image processing, and outlines the automatically optimising implementation which achieves portability of application code and efficiency of implementation on a closely coupled distributed memory parallel system. the paper includes timings for optimised and unoptimised versions of typical image processingalgorithms;it draws the main conclusion that it is possible to achieve portability with efficiency, for a specific application, by adopting a high level algebraic programming model, together with a transformation-based optimiser which reclaims the loss of efficiency which an algebraic approach traditionally entails.
暂无评论