Modern discrete GPUs support unified memory and demand paging. Automatic
management of data movement between CPU memory and GPU memory dramatically …

The proliferation of machine learning applications has promoted both CUDA Cores and
Tensor Cores' integration to meet their acceleration demands. While studies have shown …

Graphics Processing Units (GPUs) are increasingly widely used in the cloud to accelerate
compute-heavy tasks. However, GPU-compute applications stress the GPU architecture in …

Graphics Processing Units (GPUs) utilize memory hierarchy and Thread-Level Parallelism
(TLP) to tolerate off-chip memory latency, which is a significant bottleneck for memory-bound …

Graphics Processing Units (GPUs) are widely-used accelerators for data-parallel
applications. In many GPU applications, GPU memory bandwidth bottlenecks performance …

As GPUs continue to grow in popularity for accelerating demanding applications, such as
high-performance computing and machine learning, GPU architects need to deliver more …

The ability to support multitasking becomes more and more important in the development of
graphic processing unit (GPU). GPU multitasking methods are classified into three types …

Predicting performance degradation of a GPU application when it is co-located with other
applications on a spatial multitasking GPU without prior application knowledge is essential …

GPUs are becoming first-class compute citizens and increasingly support programmability-
enhancing features such as shared virtual memory and hardware cache coherence. This …

Graphics Processing Units (GPUs) use caches to provide on-chip bandwidth as a way to
address the memory wall. However, they are not always efficiently utilized for optimal GPU …