Sciweavers

IEEEPACT
2003
IEEE
14 years 5 months ago
Resolving Register Bank Conflicts for a Network Processor
This paper discusses a register bank assignment problem for a popular network processor--Intel's IXP. Due to limited data paths, the network processor has a restriction that ...
Xiaotong Zhuang, Santosh Pande
IEEEPACT
2003
IEEE
14 years 5 months ago
Miss Rate Prediction across All Program Inputs
Improving cache performance requires understanding cache behavior. However, measuring cache performance for one or two data input sets provides little insight into how cache behav...
Yutao Zhong, Steve Dropsho, Chen Ding
IEEEPACT
2003
IEEE
14 years 5 months ago
Y-Branches: When You Come to a Fork in the Road, Take It
In this paper, we study the effects of manipulating the architected direction of conditional branches. Through the use of statistical sampling, we find that about 40% of all dyna...
Nicholas J. Wang, Michael Fertig, Sanjay J. Patel
IEEEPACT
2003
IEEE
14 years 5 months ago
Initial Observations of the Simultaneous Multithreading Pentium 4 Processor
This paper analyzes an Intel Pentium 4 hyper-threading processor. The focus is to understand its performance and the underlying reasons behind that performance. Particular attenti...
Nathan Tuck, Dean M. Tullsen
IEEEPACT
2003
IEEE
14 years 5 months ago
The Impact of Resource Partitioning on SMT Processors
Simultaneous multithreading (SMT) increases processor throughput by multiplexing resources among several threads. Despite the commercial availability of SMT processors, several as...
Steven E. Raasch, Steven K. Reinhardt
IEEEPACT
2003
IEEE
14 years 5 months ago
Reducing Datapath Energy through the Isolation of Short-Lived Operands
We present a technique for reducing the power dissipation in the course of writebacks and committments in a datapath that uses a dedicated architectural register file (ARF) to hol...
Dmitry Ponomarev, Gurhan Kucuk, Oguz Ergin, Kanad ...
IEEEPACT
2003
IEEE
14 years 5 months ago
Picking Statistically Valid and Early Simulation Points
Modern architecture research relies heavily on detailed pipeline simulation. Simulating the full execution of an industry standard benchmark can take weeks to months to complete. ...
Erez Perelman, Greg Hamerly, Brad Calder
IEEEPACT
2003
IEEE
14 years 5 months ago
Design Trade-Offs in High-Throughput Coherence Controllers
Recent research shows that the high occupancy of Coherence Controllers (CCs) is a major performance bottleneck in scalable shared-memory multiprocessors. In this paper, we propose...
Anthony-Trung Nguyen, Josep Torrellas
IEEEPACT
2003
IEEE
14 years 5 months ago
Redeeming IPC as a Performance Metric for Multithreaded Programs
Recent work has shown that multithreaded workloads running in execution-driven, full-system simulation environments cannot use instructions per cycle (IPC) as a valid performance ...
Kevin M. Lepak, Harold W. Cain, Mikko H. Lipasti
IEEEPACT
2003
IEEE
14 years 5 months ago
Efficient Resource Management during Instruction Scheduling for the EPIC Architecture
Effective modeling and management of hardware resources have always been critical toward generating highly efficient code in static compilers. With Just-In-Time compilation and dy...
Dong-yuan Chen, Lixia Liu, Chen Fu, Shuxin Yang, C...