2018 · 31 citations · 83 references
EngineeringComputer EngineeringComputer ScienceComputer ArchitectureLow-power DesignParallel ComputingAccelerator ArchitectureAccelerator IntegrationGraphics Processing Unit ArchitectureGraphics Processing UnitsHardware AccelerationMemory Subsystem DesignAcceleratorsEnergy-efficient Hardware AccelerationGpu AcceleratorsGpu ArchitectureGpu MicroarchitectureMemory Hierarchy IntegrationSimt ExecutionSimt Execution Model
GPUs are bottlenecked by the off-chip communication bandwidth and its energy cost; hence near-data acceleration is particularly attractive for GPUs. Integrating the accelerators within DRAM can mitigate these bottlenecks and additionally expose them to the higher internal bandwidth of DRAM. However, such an integration is challenging, as it requires low-overhead accelerators while supporting a diverse set of applications. To enable the integration, this work leverages the approximability of GPU applications and utilizes the neural transformation, which converts diverse regions of code mainly to Multiply-Accumulate (MAC). Furthermore, to preserve the SIMT execution model of GPUs, we also propose a novel approximate MAC unit with a significantly smaller area overhead. As such, this work introduces AxRam---a novel DRAM architecture---that integrates several approximate MAC units. AxRam offers this integration without increasing the memory column pitch or modifying the internal architecture of the DRAM banks. Our results with 10 GPGPU benchmarks show that, on average, AxRam provides 2.6× speedup and 13.3× energy reduction over a baseline GPU with no acceleration. These benefits are achieved while reducing the overall DRAM system power by 26% with an area cost of merely 2.1%.
83
Sheng Li, Jung Ho Ahn, Richard Strong et al. · 2009 · 2.4K citations
Analyzing CUDA workloads using a detailed GPU simulator
Ali Bakhoda, George L. Yuan, Wilson Wai Lun Fung et al. · 2009 · 1.7K citations
Yu‐Hsin Chen, Joel Emer, Vivienne Sze · ACM SIGARCH Computer Architecture News · 2016 · 1.2K citations
A scalable processing-in-memory accelerator for parallel graph processing
Junwhan Ahn, Sungpack Hong, Sungjoo Yoo et al. · 2015 · 761 citations
Dynamic Warp Formation and Scheduling for Efficient GPU Control Flow
Wilson Wai Lun Fung, Ivan Sham, George L. Yuan et al. · 2007 · 727 citations