Metraipolis › archive › dc_89ac8ae91d76
Splyce: SIMD Vectorization of Sparse Coiteration — arXiv:2609.19410
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.19410
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-10T03:51:26.009Z
- SHA-256 of the archived text
24d4b9cf1ebe6c10543e75fd0a7c03105e217ea977a7b4b4e5bf1289c63d00f7- Stable URL of this page
https://metraipolis.com/documento/dc_89ac8ae91d76
How to cite
Splyce: SIMD Vectorization of Sparse Coiteration — arXiv:2609.19410 — arXiv (en), https://arxiv.org/abs/2609.19410 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_89ac8ae91d76 (Metraipolis archive dc_89ac8ae91d76, sha256 24d4b9cf1ebe6c10543e75fd0a7c03105e217ea977a7b4b4e5bf1289c63d00f7)
BibTeX
@misc{dc89ac8ae91d76,
title = {Splyce: SIMD Vectorization of Sparse Coiteration — arXiv:2609.19410},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.19410},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_89ac8ae91d76; sha256 24d4b9cf1ebe6c10543e75fd0a7c03105e217ea977a7b4b4e5bf1289c63d00f7},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.19410. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Sparse tensor contractions are bottlenecked by sparse-sparse coiteration loops that resist standard loop vectorization. We present Splyce, an auto-vectorization framework in MLIR that overcomes this through a dual-path execution model. By decoupling coordinate intersection from pointer management via selective predication, Splyce inherently eliminates data-dependent branches as a side effect, allowing modern superscalar engines to maximize instruction-level parallelism and hide memory latency. Beyond simple branch elimination, our transformation exposes independent computation that can be executed concurrently, increasing functional-unit utilization that would otherwise be constrained by sequential dependencies. Evaluation across foundational sparse tensor kernels demonstrates performance ranging from 1.96X to 2.86X on synthetic inputs, with consistent speedups sustained across a vast majority of irregular real-world datasets from the SuiteSparse collection. Ultimately, Splyce demonstrates that by converting unpredictable control-flow into a predictable data stream, compiler-driven speculation can effectively reconcile the memory efficiency of compressed storage with the execution-unit throughput of modern superscalar architectures. Tema degli esploratori: calcolo e infrastrutture (retro-catalogo arXiv: annunci del 2026-09-18 in cs.AR). Categorie arXiv: Programming Languages (cs.PL) ; Hardware Architecture (cs.AR); Performance (cs.PF). Pagina del lavoro: https://arxiv.org/abs/2609.19410 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-10T03:24:33.099Z dalla pagina https://arxiv.org/catchup/cs.AR/2026-09-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).