Specialized on-chip accelerators are widely used to improve the energy efficiency of
computing systems. Recent advances in memory technology have enabled near-data …

GPUs achieve high throughput and power efficiency by employing many small single
instruction multiple thread (SIMT) cores. To minimize scheduling logic and performance …

Historically, improvement in GPU performance has been tightly coupled with transistor
scaling. As Moore's Law slows down, performance of single GPUs may ultimately plateau …

This paper presents the first formal analysis of the official memory consistency model for the
NVIDIA PTX virtual ISA. Like other GPU memory models, the PTX memory model is weakly …

We present Chronos, a framework to build accelerators for applications with speculative
parallelism. These applications consist of atomic tasks, sometimes with order constraints …

Transfer learning in natural language processing (NLP) uses increasingly large models that
tackle challenging problems. Consequently, these applications are driving the requirements …

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

Cache coherence is ubiquitous in shared memory multiprocessors because it provides a
simple, high performance memory abstraction to programmers. Recent work suggests …

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 …

An unambiguous and easy-to-understand memory consistency model is crucial for ensuring
correct synchronization and guiding future design of heterogeneous systems. In a widely …