Addressing Dark Silicon Challenges with Disciplined Approximate Computing
نویسندگان
چکیده
With the power constraints of the dark silicon era looming, architectural changes to improve energy efficiency are critical. Disciplined approximate computing, in which unreliable hardware components are safely used in error-tolerant software, is a promising approach. We discuss programming models, architectures, and accelerators for approximate computing.
منابع مشابه
Jack Sampson Research Statement
Research Overview For several decades, improvements to CMOS process technologies fueled rapid growth in processor performance and throughput. Recently, however, power limitations have radically altered performance scaling for single-threaded applications, and the number of cores per processor has lagged behind the historical rate of transistor increase. This transition leaves architects with fa...
متن کاملRe-form: FPGA-powered true codesign flow for high-performance computing in the post-Moore era
Multicore scaling will end soon because of practical power limits. Dark silicon is becoming a major issue even more than the end of Moore’s law. In the post-Moore era, the energy efficiency of computing will be a major concern. FPGAs could be a key to maximizing the energy efficiency. In this paper we address severe challenges in the adoption of FPGA in HPC and describe “Re-form,” an FPGA-power...
متن کاملProcess-Variation and Soft-Error Reliability-Aware Workload Mapping with Adaptive Parallelism in SoCs
With deep technology scaling, significant design challenges are projected due to process variations, soft-errors, and dark-silicon in integrated circuits. It is well known that spatial variations in process parameters introduce unpredictability in the performance and power profiles of systems-on-chip (SoCs). By intelligently mapping applications on to the best set of available cores, process-va...
متن کاملQuality of Service Profiling and Autotuning for Energy-Aware Approximate Programming
One promising approach to energy-efficient computation, approximate computing, trades off output precision for gains in energy efficiency. Many applications can easily tolerate small errors, especially if they are handled in a disciplined manner. However, approximation introduces an inherent tradeoff between quality of service and energy efficiency. Existing approaches lack ways to quantify and...
متن کاملApproximate Accelera on for a Post Mul core Era
Approximate Accelera on for a Post Mul core Era Hadi Esmaeilzadeh Co-Chairs of the Supervisory Commi ee: Professor Doug Burger Microso Research Associate Professor Luis Ceze University of Washington Star ng in 2004, the microprocessor industry has shi ed to mul core scaling—increasing the number of cores per die each technology genera on—as its principal strategy for con nuing performance growt...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2012