{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/clblast-a-tuned-opencl-blas-library","title":"CLBlast: A Tuned OpenCL BLAS Library","arxiv_id":"1705.05249","date":"2017-05-12","proceeding":null,"authors":["Cedric Nugteren"],"abstract":"This work introduces CLBlast, an open-source BLAS library providing optimized\nOpenCL routines to accelerate dense linear algebra for a wide variety of\ndevices. It is targeted at machine learning and HPC applications and thus\nprovides a fast matrix-multiplication routine (GEMM) to accelerate the core of\nmany applications (e.g. deep learning, iterative solvers, astrophysics,\ncomputational fluid dynamics, quantum chemistry). CLBlast has five main\nadvantages over other OpenCL BLAS libraries: 1) it is optimized for and tested\non a large variety of OpenCL devices including less commonly used devices such\nas embedded and low-power GPUs, 2) it can be explicitly tuned for specific\nproblem-sizes on specific hardware platforms, 3) it can perform operations in\nhalf-precision floating-point FP16 saving bandwidth, time and energy, 4) it has\nan optional CUDA back-end, 5) and it can combine multiple operations in a\nsingle batched routine, accelerating smaller problems significantly. This paper\ndescribes the library and demonstrates the advantages of CLBlast experimentally\nfor different use-cases on a wide variety of OpenCL hardware.","url_abs":"http://arxiv.org/abs/1705.05249v2","url_pdf":"http://arxiv.org/pdf/1705.05249v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"clblast-a-tuned-opencl-blas-library","repo_url":"https://github.com/CNugteren/CLBlast","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}