This paper presents novel cache optimizations for massively parallel, throughput-oriented
architectures like GPUs. L1 data caches (L1 D-caches) are critical resources for providing …

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 …

High throughput architectures rely on high thread-level parallelism (TLP) to hide execution
latencies. In state-of-art graphics processing units (GPUs), threads are organized in a grid of …

Massively multithreaded GPUs achieve high throughput by running thousands of threads in
parallel. To fully utilize the hardware, workloads spawn work to the GPU in bulk by launching …

Modern GPUs require tens of thousands of concurrent threads to fully utilize the massive
amount of processing resources. However, thread concurrency in GPUs can be diminished …

Parallel programs consist of series of code sections with different thread-level parallelism
(TLP). As a result, it is rather common that a thread in a parallel program, such as a GPU …

Multi-application execution in Graphics Processing Units (GPUs), a promising way to utilize
GPU resources, is still challenging. Some pieces of prior work (eg, spatial multitasking) have …

For the rapid growth computation requirements in big data and artificial intelligence area,
CPU-GPU heterogeneous clusters can provide more powerful computing capacity …

Context switching is a key technique enabling preemption and time-multiplexing for CPUs.
However, for single-instruction multiple-thread (SIMT) processors such as high-end graphics …

With the unprecedented development of compute capability and extension of memory
bandwidth on modern GPUs, parallel communication and synchronization soon becomes a …