Sciweavers

2048 search results - page 39 / 410
» Optimizing pipelines for power and performance
Sort
View
VLSID
2004
IEEE
147views VLSI» more  VLSID 2004»
16 years 6 months ago
High-Performance Power Grids For Nanometer Technologies
With shrinking noise margins and increasing numbers of on-chip noise sources, power grid design has become a critical performance determinant. This paper presents an overview of r...
Sachin S. Sapatnekar
ISCAS
2003
IEEE
147views Hardware» more  ISCAS 2003»
15 years 11 months ago
Parameterized and low power DSP core for embedded systems
Conventional ASIC designs are hard to be customized. Therefore DSP core-based ASIC design has potentially large payoff. This approach not only supports improved performance but al...
Ya-Lan Tsao, Ming Hsuan Tan, Jun-Xian Teng, Shyh-J...
CODES
2006
IEEE
16 years 5 days ago
Streamroller: : automatic synthesis of prescribed throughput accelerator pipelines
In this paper, we present a methodology for designing a pipeline of accelerators for an application. The application is modeled using sequential C language with simple stylization...
Manjunath Kudlur, Kevin Fan, Scott A. Mahlke
DAC
2012
ACM
13 years 8 months ago
Equivalence checking for behaviorally synthesized pipelines
Loop pipelining is a critical transformation in behavioral synthesis. It is crucial to producing hardware designs with acceptable latency and throughput. However, it is a complex ...
Kecheng Hao, Sandip Ray, Fei Xie
ICS
1989
Tsinghua U.
15 years 10 months ago
Control flow optimization for supercomputer scalar processing
Control intensive scalar programs pose a very different challenge to highly pipelined supercomputers than vectorizable numeric applications. Function call/return and branch instru...
Pohua P. Chang, Wen-mei W. Hwu