Abstract
Embedded systems have stringent design constraints, which has necessitated much prior research focus on optimizing energy consumption and/or performance. Since embedded systems typically have fewer cooling options, rising temperature, and thus temperature optimization, is an emergent concern. Most embedded systems only dissipate heat by passive convection, due to the absence of dedicated thermal management hardware mechanisms. The embedded system’s temperature not only affects the system’s reliability, but can also affect the performance, power, and cost. Thus, embedded systems require efficient thermal management techniques. However, thermal management can conflict with other optimization objectives, such as execution time and energy consumption. In this paper, we focus on managing the temperature using a synergy of cache optimization and dynamic frequency scaling, while also optimizing the execution time and energy consumption. This paper provides new insights on the impact of cache parameters on efficient temperature-aware cache tuning heuristics. In addition, we present temperature-aware phase-based tuning, TaPT, which determines Pareto optimal clock frequency and cache configurations for fine-grained execution time, energy, and temperature tradeoffs. TaPT enables autonomous system optimization and also allows designers to specify temperature constraints and optimization priorities. Experiments show that TaPT can effectively reduce execution time, energy, and temperature, while imposing minimal hardware overhead.
Highlights
Embedded systems have become ubiquitous over the past few years, and with the emergence and growth of the Internet of Things, embedded systems are expected to become even more pervasive
We focus on the cache for phase-based tuning, and the clock frequency for dynamic thermal management (DTM), using dynamic frequency scaling (DFS)
Temperature-Aware Phase-based Tuning (TaPT)’s default setting of S (EDP prioritization), which represents a system with no designer-specified priority or temperature threshold
Summary
Embedded systems have become ubiquitous over the past few years, and with the emergence and growth of the Internet of Things, embedded systems are expected to become even more pervasive. Researchers have focused on effective optimization techniques for optimizing embedded systems’ energy consumption, since these systems typically have stringent resource and design constraints. These constraints include form factor, battery capacity, cost, real-time deadlines, etc., and pose significant challenges to embedded system optimization. Sherwood et al [18] studied applications’ time varying behaviors using SPEC 95 benchmarks, and showed that applications have periodic patterns and exhibit phase-based behavior with respect to several execution statistics (e.g., cache miss rates, branch mispredicts, IPC, etc.) Balasubramonian et al [19] used cache miss rates, cycles per instruction (CPI), and branch frequency characteristics to detect changes in application characteristics for cache tuning, and found that these characteristics were effective for phase classification
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call