qscores_trading_dark_silicon_for_scalable_energy_efficiencyqscores交易黑硅的可伸缩的能源效率.pdf

QSCORES : Trading Dark Silicon for Scalable Energy Efficiency with Quasi-Specific Cores Ganesh Venkatesh , Jack Sampson, Nathan Goulding-Hotta, Sravanthi Kota Venkata , Michael Bedford Taylor and Steven Swanson Department of Computer Science and Engineering University of California, San Diego ABSTRACT General Terms Transistor density continues to increase exponentially, but power Design, Experimentation, Measurement dissipation per transistor is improving only slightly with each gen- eration of Moore’s law. Given the constant chip-level power bud- Keywords gets, this exponentially decreases the percentage of transistors that QSCORE, Conservation Core, Merging, Dark Silicon, Utilization can switch at full frequency with each technology generation. Hence, Wall, Heterogeneous Many-Core, Specialization while the transistor budget continues to increase exponentially, the power budget has become the dominant limiting factor in processor 1. INTRODUCTION design. In this regime, utilizing transistors to design specialized cores that optimize energy-per-computation becomes an effective Although transistor density continues to scale, nearly constant approach to improve system performance. per-transistor power and fixed chip-level power budget place tight To trade transist