Sciweavers

10805 search results - page 37 / 2161
» Mapping Computation with No Memory
Sort
View
CONCURRENCY
2002
112views more  CONCURRENCY 2002»
13 years 8 months ago
An analysis of VI Architecture primitives in support of parallel and distributed communication
We present the results of a detailed study of the Virtual Interface (VI) paradigm as a communication foundation for a distributed computing environment. Using Active Messages and ...
Andrew Begel, Philip Buonadonna, David E. Culler, ...
IPPS
2010
IEEE
13 years 6 months ago
Optimization of linked list prefix computations on multithreaded GPUs using CUDA
We present a number of optimization techniques to compute prefix sums on linked lists and implement them on multithreaded GPUs using CUDA. Prefix computations on linked structures ...
Zheng Wei, Joseph JáJá
ICCAD
2007
IEEE
123views Hardware» more  ICCAD 2007»
14 years 5 months ago
Mapping model with inter-array memory sharing for multidimensional signal processing
Abstract – The storage requirements in data-intensive signal processing systems (including applications in video and image processing, artificial vision, medical imaging, real-t...
Ilie I. Luican, Hongwei Zhu, Florin Balasa
ICASSP
2008
IEEE
14 years 3 months ago
Efficient assignment algorithm for mapping multidimensional signals into the physical memory
The storage requirements in data-intensive multidimensional signal processing systems have a significant impact on the system performance as well as on essential design parameter...
Ilie I. Luican, Hongwei Zhu, Florin Balasa
FPGA
1998
ACM
153views FPGA» more  FPGA 1998»
14 years 29 days ago
SMAP: Heterogeneous Technology Mapping for Area Reduction in FPGAs with Embedded Memory Arrays
It has become clear that large embedded con gurable memory arrays will be essential in future FPGAs. Embedded arrays provide high-density high-speed implementations of the storage...
Steven J. E. Wilton