Sciweavers

176 search results - page 6 / 36
» Performance improvement with circuit-level speculation
Sort
View
HPCA
2002
IEEE
14 years 8 months ago
Improving Value Communication for Thread-Level Speculation
Thread-Level Speculation (TLS) allows us to automatically parallelize general-purpose programs by supporting parallel execution of threads that might not actually be independent. ...
J. Gregory Steffan, Christopher B. Colohan, Antoni...
JILP
2000
79views more  JILP 2000»
13 years 8 months ago
A Comparative Survey of Load Speculation Architectures
Load latency remains a signi cant bottleneck in dynamically scheduled pipelined processors. Load speculation techniques have been proposed to reduce this latency. Dependence Predi...
Brad Calder, Glenn Reinman
ISCA
1999
IEEE
124views Hardware» more  ISCA 1999»
14 years 21 days ago
Speculation Techniques for Improving Load Related Instruction Scheduling
State of the art microprocessors achieve high performance by executing multiple instructions per cycle. In an out-oforder engine, the instruction scheduler is responsible for disp...
Adi Yoaz, Mattan Erez, Ronny Ronen, Stéphan...
IPPS
2006
IEEE
14 years 2 months ago
Improving cache locality for thread-level speculation
With the advent of chip-multiprocessors (CMPs), Thread-Level Speculation (TLS) remains a promising technique for exploiting this highly multithreaded hardware to improve the perfo...
Stanley L. C. Fung, J. Gregory Steffan
ASYNC
1997
IEEE
66views Hardware» more  ASYNC 1997»
14 years 19 days ago
Speculative Completion for the Design of High-Performance Asynchronous Dynamic Adders
This paper presents an in-depth case study in highperformance asynchronous adder design. A recent method, called “speculative completion”, is used. This method uses single-rai...
Steven M. Nowick, Kenneth Y. Yun, Ayoob E. Dooply,...