Transaction level statistical analysis for efficient micro-architectural power and performance studies

Copty, Eman; Kamhi, Gila; Novakovsky, Sasha

doi:10.1145/2024724.2024809

Cited by 4 publications

(1 citation statement)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For power estimation, popular approaches are to leverage high-level state-based models, to develop macro models for micro-architectural functional blocks [8,27], to use detailed simulation-based modeling frameworks [5,16], or to support estimation at a range of abstraction [21]. Simulation-based methods utilize activity information gathered from cycle accurate performance simulators to estimate power consumption.…”

Section: Related Workmentioning

confidence: 99%

Automated, retargetable back-annotation for host compiled performance and power modeling

Chakravarty

Zhao

Gerstlauer

2013

2013 International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS)

View full text Add to dashboard Cite

With traditional cycle-accurate or instruction-set simulations of processors often being too slow, host-compiled or source-level software execution approaches have recently become popular. Such high-level simulations can achieve order of magnitude speedups, but approaches that can achieve highly accurate characterization of both power and performance metrics are lacking. In this paper, we propose a novel host-compiled simulation approach that provides close to cycle-accurate estimation of energy and timing metrics in a retargetable manner, using flexible, architecture description language (ADL) based reference models. Our automated flow considers typical front-and back-end optimizations by working at the compiler-generated intermediate representation (IR). Path-dependent execution effects are accurately captured through pairwise characterization and backannotation of basic code blocks with all possible predecessors. Results from applying our approach to PowerPC targets running various benchmark suites show that close to native average speeds of 2000 MIPS at more than 98% timing and energy accuracy can be achieved.

show abstract