Abstract:Energy-efficiency has become a major challenge in modern computer systems. To address this challenge, candidate systems increasingly integrate heterogeneous cores in order to satisfy diverse computation requirements by selecting cores with suitable features. In particular, single-ISA heterogeneous multicore processors such as ARM <a class="link-external link-http" href="http://big.LITTLE" rel="external noopener nofollow">this http URL</a> have become very attractive since they offer good opportunities in terms of performance and power consumption trade-off. While existing works already showed that this feature can improve system energy-efficiency, further gains are possible by generalizing the principle to higher levels of heterogeneity. The present paper aims to explore these gains by considering single-ISA heterogeneous multicore architectures including three different types of cores. For this purpose, we use the Samsung Exynos Octa 5422 chip as baseline architecture. Then, we model and evaluate Cortex A7, A9, and A15 cores using the gem5 simulation framework coupled to McPAT for power estimation. We demonstrate that varying the level of heterogeneity as well as the different core ratio can lead to up to 2.3x gains in energy efficiency and up to 1.5x in performance. This study further provides insights on the impact of workload nature on performance/energy trade-off and draws recommendations concerning suitable architecture configurations. This contributes in fine to guide future research towards dynamically reconfigurable HSAs in which some cores/clusters can be disabled momentarily so as to optimize certain metrics such as energy efficiency. This is of particular interest when dealing with quality-tunable algorithms in which accuracy can be then traded for compute effort, thereby enabling to use only those cores that provide the best energy-efficiency for the chosen algorithm.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is how to find the optimal balance between performance and energy consumption in a heterogeneous multi - core architecture. Specifically, the paper focuses on exploring the potential gains in system performance and energy efficiency by introducing a higher level of heterogeneity (i.e., integrating three different types of processor cores under a single Instruction Set Architecture (ISA)). The paper uses the Samsung Exynos Octa 5422 chip as the baseline architecture and models and evaluates the Cortex A7, A9, and A15 cores through the gem5 cycle - approximate simulation framework and the McPAT power consumption evaluation tool. The study also considers the influence of application characteristics in the Rodinia benchmark suite, such as memory bandwidth, computational complexity, and communication patterns, on the performance/energy trade - off, and proposes suggestions suitable for architecture configurations. These studies are helpful for guiding future research on Dynamic Reconfigurable Heterogeneous System Architectures (HSA), especially when dealing with quality - adjustable algorithms, being able to dynamically select the most energy - efficient core according to the algorithm requirements.

Exploration of Performance and Energy Trade-offs for Heterogeneous Multicore Architectures

Exploiting Heterogeneity For Energy Efficiency In Chip Multiprocessors

Optimizing the Energy Efficiency of Power Supply in Heterogeneous Multicore Chips with Integrated Switched-Capacitor Converters

Evaluation of hybrid run-time power models for the ARM big.LITTLE architecture

High-Performance Simultaneous Multiprocessing for Heterogeneous System-on-Chip

Performance Evaluation and Optimization of Inter-cores Communication for Heterogeneous Multi-core Processor Unit

Energy Efficiency Analysis of Heterogeneous Platforms: Early Experiences.

Collaborative Heterogeneity-Aware OS Scheduler for Asymmetric Multicore Processors

Energy-Efficient Resource Utilization for Heterogeneous Embedded Computing Systems

Energy Efficient Job Scheduling in Single-Isa Heterogeneous Chip-Multiprocessors

Bi-Directional Timing-Power Optimisation on Heterogeneous Multi-Core Architectures

An Evaluation of Per-Chip Nonuniform Frequency Scaling on Multicores.

Research of the Heterogeneous Multi-Core Processor Architecture Design

Cross-architecture Prediction Based Scheduling for Energy Efficient Execution on Single-Isa Heterogeneous Chip-Multiprocessors.

Heterogeneous systems for energy efficient scientific computing

Parallelizing Workload Execution in Embedded and High-Performance Heterogeneous Systems

STEER: Asymmetry-aware Energy Efficient Task Scheduler for Cluster-based Multicore Architectures

Design Space Exploration for Device and Architectural Heterogeneity in Chip-Multiprocessors

Efficient Scheduling Mechanism for Performance-Heterogeneous Multi-core Processor

Heterogeneous Architecture Design with Emerging 3D and Non-Volatile Memory Technologies.

The Design and Implementation of Heterogeneous Multicore Architecture in a Processor-Configurable Environment