We present a novel loop transformation technique, particularly well suited for optimizing embedded compilers, where an increase in compilation time is acceptable in exchange for s...
Graphics processing units (GPUs) provide a low cost platform for accelerating high performance computations. The introduction of new programming languages, such as CUDA and OpenCL...
Amir Hormati, Mehrzad Samadi, Mark Woh, Trevor N. ...
The widening gap between CPU and memory speed has made caches an integral feature of modern highperformance processors. The high degree of configurability of cache memory can requ...
Rahman Hassan, Antony Harris, Nigel P. Topham, Ari...
Abstract. In this paper, we consider the problem of generating optimized, executable control code from high-level, symbolic specifications. In particular, we construct symbolic co...
Handling large amounts of data, such as large image databases, requires the use of approximate nearest neighbor search techniques. Recently, Hamming embedding methods such as spec...