The demands of high-performance computing (HPC) and machine learning (ML) workloads
have resulted in the rapid architectural evolution of GPUs over the last decade. The growing …

Recent heterogeneous architectures have trended toward tighter integration and shared
memory largely due to the efficient communication and programmability enabled by this …

Prior work on GPU cache coherence has shown that simple hardware-or software-based
protocols can be more than sufficient. However, in recent years, features such as multi-chip …

Heterogeneous CPU-FPGA systems are evolving towards tighter integration between CPUs
and FPGAs for improved performance and energy efficiency. At the same time …

Running experiments in modern computer architecture simulators can be a difficult and error-
prone endeavor. Users must track many configurations, components and outputs between …

This paper presents ALTIS, a benchmark suite for modern GPGPU computing. Previous
benchmark suites such as Rodinia and SHOC have served the research community well, but …

Deterministic execution for GPUs is a desirable property as it helps with debuggability and
reproducibility. It is also important for safety regulations, as safety critical workloads are …

Newer use cases of GPU (Graphics Processing Unit) computing, eg, graph analytics, look
less like traditional bulk-synchronous GPU programs. To cater to the needs of emerging …

Today, GPUs are widely used as coprocessors/accelerators in High-Performance
Heterogeneous Computing due to their many advantages. However, many researches …

GPUs have evolved from providing highly-constrained programmability for a single kernel to
using pre-emption to ensure independent forward progress for concurrently executing …