Sciweavers

901 search results - page 60 / 181
» Hiding Communication Latency in Data Parallel Applications
Sort
View
HPDC
2002
IEEE
14 years 1 months ago
Multigrain Parallelism for Eigenvalue Computations on Networks of Clusters
Clusters of workstations have become a cost-effective means of performing scientific computations. However, large network latencies, resource sharing, and heterogeneity found in ...
James R. McCombs, Andreas Stathopoulos
ISPD
2003
ACM
132views Hardware» more  ISPD 2003»
14 years 1 months ago
Architecture and synthesis for multi-cycle communication
For multi-gigahertz designs in nanometer technologies, data transfers on global interconnects take multiple clock cycles. In this paper, we propose a regular distributed register ...
Jason Cong, Yiping Fan, Xun Yang, Zhiru Zhang
IEEEPACT
1999
IEEE
14 years 10 days ago
Memory System Support for Image Processing
Image processing applications tend to access their data non-sequentially and reuse that data infrequently. As a result, they tend to perform poorly on conventional memory systems ...
Lixin Zhang, John B. Carter, Wilson C. Hsieh, Sall...
VISUALIZATION
2003
IEEE
14 years 1 months ago
Planet-Sized Batched Dynamic Adaptive Meshes (P-BDAM)
We describe an efficient technique for out-of-core management and interactive rendering of planet sized textured terrain surfaces. The technique, called P-Batched Dynamic Adaptiv...
Paolo Cignoni, Fabio Ganovelli, Enrico Gobbetti, F...
ISCAS
2005
IEEE
144views Hardware» more  ISCAS 2005»
14 years 1 months ago
Event-based imaging with active illumination in sensor networks
— We discuss a distributed imaging architecture with active illumination for sensor network applications. An event-based CMOS imager is employed at the sensor level, to convert l...
Eugenio Culurciello, Thiago Teixeira, Andreas G. A...