GPU

Analyzing GCN Aggregation on GPU
Graph convolutional neural networks (GCNs) are emerging neural networks for graph structures that include large features associated …
PDF Cite Project Project DOI
Analyzing GCN Aggregation on GPU
GhostLeg: Selective Memory Coalescing for Secure GPU Architecture
Architectural considerations for secure executions are getting more critical for GPU since popular security applications and libraries …
PDF Cite Project Project DOI
GhostLeg: Selective Memory Coalescing for Secure GPU Architecture
Analyzing Characteristics of Memory Timing Side-Channels in GPU
최근 연구에서 GPU에서 수행하는 AES와 RSA와 같은 암호 알고리즘에서 GPU의 고유한 메모리 접합 구조에 따른 메모리 접근 시간의 차이를 이용하여 공격자가 암호키를 복원할 수 있음이 밝혀졌다. 이는 GPU 캐시 구조 및 메모리 접합 방식 …
PDF Cite Project Project Slides Slide Show
Analyzing Characteristics of Memory Timing Side-Channels in GPU
Analyzing Data Cache Performance of Modern GPU Architecture
Graphics processing unit (GPU)는 그래픽 어플리케이션의 처리 뿐만 아니라 최근 machine learning, big data analytics 등의 대규모 병렬처리를 요구하는 어플리케이션의 처리에 널리 사용되고 있다. …
PDF Cite Project Slides Slide Show
Analyzing Data Cache Performance of Modern GPU Architecture
Analyzing the Performance of GCN Inferences with respect to Sparsity of Graph Features
Graph Convolutional Neural Network (GCN)은 그래프 구조를 이용한 인공 신경망 중의 하나로 소셜 네트워크 분석 및 소비자 성향 분석 등 여러 응용 분야에 이용될 수 있다. GCN 커널은 크게 aggregation …
PDF Cite Project Project Slides Slide Show
Analyzing the Performance of GCN Inferences with respect to Sparsity of Graph Features
Revealing Characteristics of GCN Inference Models Using a GPU Profiler
Graph Convolutional Neural Network (GCN)은 그래프 구조를 이용한 인공 신경망 (Graph Neural Network, GNN) 중의 하나로서 소셜 네트워크 분석 및 소비자 성향 분석, 추천 시스템 등의 여러 응 …
PDF Cite Project Project Slides Slide Show
Revealing Characteristics of GCN Inference Models Using a GPU Profiler
Hi-End: Hierarchical, Endurance-Aware STT-MRAM-Based Register File for Energy-Efficient GPUs
Modern Graphics Processing Units (GPUs) require large hardware resources for massive parallel thread executions. In particular, modern …
PDF Cite Project DOI
Hi-End: Hierarchical, Endurance-Aware STT-MRAM-Based Register File for Energy-Efficient GPUs
Linebacker: Preserving Victim Cache Lines in Idle Register Files of GPUs
Modern GPUs suffer from cache contention due to the limited cache size that is shared across tens of concurrently running warps. To …
PDF Cite Project DOI Custom Link
Linebacker: Preserving Victim Cache Lines in Idle Register Files of GPUs
CTA-Aware Prefetching and Scheduling for GPU
Albeit GPUs are supposed to be tolerant to long latency of data fetch operation, we observe that L1 cache misses occur in a bursty …
PDF Cite Project Slides DOI Slide Show
CTA-Aware Prefetching and Scheduling for GPU
Access Pattern-Aware Cache Management for Improving Data Utilization in GPU
Long latency of memory operation is a prominent performance bottleneck in graphics processing units (GPUs). The small data cache that …
PDF Cite Project Slides DOI Slide Show
Access Pattern-Aware Cache Management for Improving Data Utilization in GPU