ORNL researcher Sparsh Mittal has authored a new paper entitled “A Survey Of Techniques for Architecting and Managing Asymmetric Multicore Processors.” Now accepted for ACM Computing Surveys 2015, the document reviews nearly 125 papers.
Modern computing systems have become highly diverse in terms of their workloads, usage pattern, scale and optimization objective. Hence, even a highly-optimized “monolithic core” cannot simultaneously meet such diverse and often conflicting requirements. To address this challenge, asymmetric multicore processors have been proposed which feature cores of different types (e.g. big and LITTLE) in the same processor. Qualcomm Snapdragon 810, Samsung Exynos 5 Octa and Nvidia Tegra X1 are some examples of asymmetric multicores.