The Paper
Halide: a language and compiler for optimizing parallelism, locality, and recomputation in image processing pipelines
Paper Link
https://people.csail.mit.edu/jrk/halide-pldi13.pdf
Format
We start at 6:10, don't be late!
The discussion lasts for about 1 to 1.5 hours, depending upon the paper.
Read the paper (done before you arrive)
Introductions (name, and background)
First impressions (1-2 minutes this is what I thought)
Structured review (we move through the paper in order, everyone gets a chance to ask questions, offer comments, and raise concerns)
Free form discussion
Nominate and vote on the next paper
Abstract
Image processing pipelines combine the challenges of stencil computations and stream programs. They are composed of large graphs of different stencil stages, as well as complex reductions, and stages with global or data-dependent access patterns. Because of their complex structure, the performance difference between a naive implementation of a pipeline and an optimized one is often an order of magnitude. Efficient implementations require optimization of both parallelism and locality, but due to the nature of stencils, there is a fundamental tension between parallelism, locality, and introducing redundant recomputation of shared values.
We present a systematic model of the tradeoff space fundamental to stencil pipelines, a schedule representation which describes concrete points in this space for each stage in an image processing pipeline, and an optimizing compiler for the Halide image processing language that synthesizes high performance implementations from a Halide algorithm and a schedule. Combining this compiler with stochastic search over the space of schedules enables terse, composable programs to achieve state-of-the-art performance on a wide range of real image processing pipelines, and across different hardware architectures, including multicores with SIMD, and heterogeneous CPU+GPU execution. From simple Halide programs written in a few hours, we demonstrate performance up to 5x faster than hand-tuned C, intrinsics, and CUDA implementations optimized by experts over weeks or months, for image processing applications beyond the reach of past automatic compilers.