A survey of methods for analyzing and improving GPU energy efficiency
Recent years have witnessed phenomenal growth in the computational capabilities and
applications of GPUs. However, this trend has also led to a dramatic increase in their power …
applications of GPUs. However, this trend has also led to a dramatic increase in their power …
Squeezedet: Unified, small, low power fully convolutional neural networks for real-time object detection for autonomous driving
Object detection is a crucial task for autonomous driving. In addition to requiring high
accuracy to ensure safety, object detection for autonomous driving also requires real-time …
accuracy to ensure safety, object detection for autonomous driving also requires real-time …
Ultra-performance Pascal GPU and NVLink interconnect
D Foley, J Danskin - IEEE Micro, 2017 - ieeexplore.ieee.org
This article introduces Nvidia's high-performance Pascal GPU. GP100 features in-package
high-bandwidth memory, support for efficient FP16 operations, unified memory, and …
high-bandwidth memory, support for efficient FP16 operations, unified memory, and …
A survey of techniques for architecting and managing GPU register file
S Mittal - IEEE Transactions on Parallel and Distributed …, 2016 - ieeexplore.ieee.org
To support their massively-multithreaded architecture, GPUs use very large register file (RF)
which has a capacity higher than even L1 and L2 caches. In total contrast, traditional CPUs …
which has a capacity higher than even L1 and L2 caches. In total contrast, traditional CPUs …
Adaptive cache management for energy-efficient GPU computing
With the SIMT execution model, GPUs can hide memory latency through massive
multithreading for many applications that have regular memory access patterns. To support …
multithreading for many applications that have regular memory access patterns. To support …
Scaling the power wall: a path to exascale
Modern scientific discovery is driven by an insatiable demand for computing performance.
The HPC community is targeting development of supercomputers able to sustain 1 ExaFlops …
The HPC community is targeting development of supercomputers able to sustain 1 ExaFlops …
Coordinated static and dynamic cache bypassing for GPUs
The massive parallel architecture enables graphics processing units (GPUs) to boost
performance for a wide range of applications. Initially, GPUs only employ scratchpad …
performance for a wide range of applications. Initially, GPUs only employ scratchpad …
A case for core-assisted bottleneck acceleration in GPUs: enabling flexible data compression with assist warps
Modern Graphics Processing Units (GPUs) are well provisioned to support the concurrent
execution of thousands of threads. Unfortunately, different bottlenecks during execution and …
execution of thousands of threads. Unfortunately, different bottlenecks during execution and …
Mascar: Speeding up GPU warps by reducing memory pitstops
With the prevalence of GPUs as throughput engines for data parallel workloads, the
landscape of GPU computing is changing significantly. Non-graphics workloads with high …
landscape of GPU computing is changing significantly. Non-graphics workloads with high …
Zorua: A holistic approach to resource virtualization in GPUs
This paper introduces a new resource virtualization framework, Zorua, that decouples the
programmer-specified resource usage of a GPU application from the actual allocation in the …
programmer-specified resource usage of a GPU application from the actual allocation in the …