2012 · 241 citations · 18 references
EngineeringComputer EngineeringComputer ScienceComputer ArchitectureField-programmable Gate ArraysLow-power DesignAccelerator ArchitectureField-programmable Gate Array ArchitectureFpga ArchitectureGraphics Processing UnitsHardware AccelerationPower OptimizationAccelerator MicroarchitectureAcceleratorsApplication-specific Accelerator DesignEnergyEnergy-efficient Hardware AccelerationEnergy-efficient Multicore DesignField-programmable Gate Arrays (Fpgas)Power-efficient Accelerator Design
The rise of multicores, GPUs, and FPGAs has created a vast design space with widely varying performance and energy outcomes across accelerators, application domains, and use cases. This paper evaluates sliding‑window applications on FPGAs, GPUs, and multicores. For each device, the authors propose optimization strategies and identify use cases where it performs best. Results show FPGAs deliver up to 11× speedup over GPUs and 57× over multicores while consuming far less energy.
With the emergence of accelerator devices such as multicores, graphics-processing units (GPUs), and field-programmable gate arrays (FPGAs), application designers are confronted with the problem of searching a huge design space that has been shown to have widely varying performance and energy metrics for different accelerators, different application domains, and different use cases. To address this problem, numerous studies have evaluated specific applications across different accelerators. In this paper, we analyze an important domain of applications, referred to as sliding-window applications, when executing on FPGAs, GPUs, and multicores. For each device, we present optimization strategies and analyze use cases where each device is most effective. The results show that FPGAs can achieve speedup of up to 11x and 57x compared to GPUs and multicores, respectively, while also using orders of magnitude less energy.
18
John D. Owens, Michael Houston, David Luebke et al. · Proceedings of the IEEE · 2008 · 1.4K citations
John D. Owens, Michael Houston, David Luebke et al. · Proceedings of the IEEE · 2008 · 1.1K citations
Aaftab Munshi · 2009 · 644 citations
Accelerating Compute-Intensive Applications with GPUs and FPGAs
Shuai Che, Jie Li, Jeremy W. Sheaffer et al. · 2008 · 317 citations