Heterogeneous multi-core processors are attracted by the media processing applications due to their capability of drawing strengths of different cores to improve the overall performance. However, the data transfer bottlenecks and limitations in the task allocation due to the accelerator-incompatible operations prevents us from gaining full potential of the heterogeneous multi-core processors. T...

In this work we introduce a history-aware, resourcebased dynamic (or simply HARD) scheduler for heterogeneous CMPs. HARD relies on recording application resource utilization and throughput to adaptively change cores for applications during runtime. We show that HARD can be configured to achieve both performance and power improvements. We compare HARD to a complexity-based static scheduler and s...

We present an overview of the Synchroscalar single-chip, multi-core processor. Through the design of Synchroscalar, we find that high energy efficiency and low complexity can be attained through parallelization. The importance of adequate inter-core interconnect is also demonstrated. We discuss the impact of having multiple frequency and voltage domains on chip to reduce the power consumption w...

We analyze two parallel finite element implementations of the 2D time-dependent advection diffusion problem, one for multi-core clusters and one for CUDA-enabled GPUs, and compare their performances in terms of time and energy consumption. The parallel CUDA-enabled GPU implementation was derived from the multi-core cluster version. Our experimental results show that a desktop machine with a sin...

The F.A.S.T. model for microscopic simulation of pedestrians was formulated with the idea of parallelizability in mind, but so far it was never demonstrated, if it can in fact be implemented efficiently for execution on a multi-core or multi-CPU system. In this contribution results are given on computation times for the F.A.S.T. model on an eight-core PC. PTV AG Karlsruhe www.ptvag.com Tobias K...

Heterogeneous multi-core processors are at tracted by the media processing applications due to their capability ofdrawing strengths ofdifferent cores to improve the overall performance. However, the data transfer bottle necks and limitations in the task allocation have preventedus from gainingfull potential of the heterogeneous multi-core processors. This paper presents a task allocation method...

Original scientific paper Requirements of digital libraries for computing power, as well as needs of smaller users for automated document processing and digitization are growing, making grid environments and multi-core systems a preferred platform for optical character recognition applications. This paper examines character recognition performance on the CRO-NGI grid and on two multi-core syste...

A graphics workstation should offer both a wide variety of 20 and 3D real­ time display functions as well as a programmable parallel-processing capacity for large processing tasks. A system concept is proposed that meets these require­ ments by offering a multi-processor configuration with general-purpose pro­ grammable processors, enhanced with specific logic that can perform for each node a l...