Cloud Computing is an emerging computing paradigm. It shares massively scalable, elastic
resources (eg, data, calculations, and services) transparently among the users over a …

Futhark is a purely functional data-parallel array language that offers a machine-neutral
programming model and an optimising compiler that generates OpenCL code for GPUs …

Computers have become increasingly complex with the emergence of heterogeneous
hardware combining multicore CPUs and GPUs. These parallel systems exhibit tremendous …

Writing high-performance GPU implementations of graph algorithms can be challenging. In
this paper, we argue that three optimizations called throughput optimizations are key to high …

Purely functional, embedded array programs are a good match for SIMD hardware, such as
GPUs. However, the naive compilation of such programs quickly leads to both code …

Compilation techniques for nested-parallel applications that can adapt to hardware and
dataset characteristics are vital for unlocking the power of modern hardware. This paper …

GPUs have been proven effective for structured applications that map well to the rigid 1D-3D
grid of threads in modern bulk synchronous parallel (BSP) programming languages …

Recent developments in GPU execution models and architectures have introduced dynamic
parallelism to facilitate the execution of irregular applications where control flow and …

Machine Learning graphs (or models) can be challenging or impossible to train when either
devices have limited memory, or the models are large. Splitting the model graph across …

GPUs lack fundamental support for data-dependent parallelism and synchronization. While
CUDA Dynamic Parallelism signals progress in this direction, many limitations and …