Sapha
As multi-core processors became common, heterogeneity among cores on a single chip created a need to assign application threads to cores so that a thread’s resource needs match resource availability at its core. Sapha designed, implemented, and evaluated automatic thread-to-core assignment techniques for heterogeneous multi-core processors to improve their utilization without requiring hand-built representative input sets.
This project concluded during the lab’s time at Iowa State University.
Publications
- TOCE '12 · Frances: A Tool For Understanding Computer Architecture and Assembly Language
- Sondag PhD Thesis · Phase-based tuning: better utilized performance asymmetric multicores
- CGO '11 · Phase-based Tuning for Better Utilization of Performance-Asymmetric Multicore Processors
- CCSC '11 · Frances-A: A Tool For Architecture Level Software Visualization
- SIGCSE '10 · Frances: A Tool For Understanding Code Generation
- RTSS '10 · A More Precise Abstract Domain for Multi-level Caches for Tighter WCET Analysis
- CCSC '10 · Connecting High-Level Programming Constructs to Assembly Language Using Frances
- Sondag MS Thesis · Phase-based tuning for better utilized performance-asymmetric multicores
- IW-MSE '09 · Phase-guided Thread-to-core Assignment for Improved Utilization of Performance-Asymmetric Multi-Core Processors
- PLOS '07 · Predictive Thread-to-Core Assignment on a Heterogeneous Multi-core Processor