Tiramisu: A Polyhedral Compiler for Expressing Fast and Portable Code
By Riyadh Baghdadi, Jessica Ray, Malek Ben Romdhane, Emanuele Del Sozzo, Abdurrahman Akkas, Yunming Zhang, Patricia Suriana, Shoaib Kamil, Saman P. Amarasinghe. IEEE/ACM International Symposium on Code Generation and Optimization, 2018
314 citations recorded by the source corpus.
This paper introduces Tiramisu, a polyhedral framework designed to generate high performance code for multiple platforms including multicores, GPUs, and distributed machines. Tiramisu introduces a scheduling language with novel commands to explicitly manage the complexities that arise when targeting these systems. The framework is designed for the areas of image processing, stencils, linear algebra and deep learning. Tiramisu has two main features: it relies on a flexible representation based on the polyhedral model and it has a rich scheduling language allowing fine-grained control of optimizations. Tiramisu uses a four-level intermediate representation that allows full separation between the algorithms, loop transformations, data layouts, and communication. This separation simplifies targeting multiple hardware architectures with the same algorithm. We evaluate Tiramisu by writing a set of image processing, deep learning, and linear algebra benchmarks and compare them with state-of-the-art compilers and hand-tuned libraries. We show that Tiramisu matches or outperforms existing compilers and libraries on different hardware architectures, including multicore CPUs, GPUs, and distributed machines.
Subfields: Hardware and Architecture
View Paper: https://www.semanticscholar.org/paper/8c7310477fd027193cd040288f0aa9824c80b91f
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Hardware and Architecture
- Albert Cohen
- River Riddle
- Uday Bondhugula
- Chris Lattner
- Tatiana Shpeisman
- Nicolas Vasilache
- Oleksandr Zinenko
- Andy Davis
Citations in the corpus
- Kernel Operations on the GPU, with Autodiff, without Memory Overflows (2020, background): “rt performances, domain-specic compilers have thus been proposed for image processing and deep learning: see for instance the Halide (Ragan-Kelley et al., 2017), TVM (Chen et al., 2018) or Tiramisu (Baghdadi et al., 2019) libraries. These frameworks are ambitious, but none of them really suits the n”
- MLIR: A Compiler Infrastructure for the End of Moore's Law (2020, background): “…metaprogramming [19, 41, 5, 14], and fully automatic flows such as PolyMage [28], Tensor Com-prehensions [52], Stripe [58], Diesel [16], Tiramisu [4] and their underlying polyhedral compilation techniques [17, 54, 8, 55] could co-exist as different code generation paths within an MLIR-based…”