Kernel Tuning Toolkit
Kernel Tuning Toolkit (KTT) is an autotuning framework for CUDA, OpenCL and Vulkan kernels. KTT provides advanced autotuning features such as support for both dynamic (online) and offline tuning, and an ability to tune multiple kernels together with shared tuning parameters. Furthermore, it offers c...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-05-01
|
| Series: | SoftwareX |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S235271102300081X |
| _version_ | 1827940587467177984 |
|---|---|
| author | Filip Petrovič Jiří Filipovič |
| author_facet | Filip Petrovič Jiří Filipovič |
| author_sort | Filip Petrovič |
| collection | DOAJ |
| description | Kernel Tuning Toolkit (KTT) is an autotuning framework for CUDA, OpenCL and Vulkan kernels. KTT provides advanced autotuning features such as support for both dynamic (online) and offline tuning, and an ability to tune multiple kernels together with shared tuning parameters. Furthermore, it offers customization features that make integration into larger software suites possible. The framework handles all major steps required for autotuning implementation, including configuration space creation and exploration, kernel code execution and output validation. The public API is available natively in C++ and via bindings in Python. |
| first_indexed | 2024-03-13T09:11:34Z |
| format | Article |
| id | doaj.art-ff6478ae80e94125879b020f890f6411 |
| institution | Directory Open Access Journal |
| issn | 2352-7110 |
| language | English |
| last_indexed | 2024-03-13T09:11:34Z |
| publishDate | 2023-05-01 |
| publisher | Elsevier |
| record_format | Article |
| series | SoftwareX |
| spelling | doaj.art-ff6478ae80e94125879b020f890f64112023-05-27T04:25:57ZengElsevierSoftwareX2352-71102023-05-0122101385Kernel Tuning ToolkitFilip Petrovič0Jiří Filipovič1Masaryk University, Brno, Czech RepublicCorresponding author.; Masaryk University, Brno, Czech RepublicKernel Tuning Toolkit (KTT) is an autotuning framework for CUDA, OpenCL and Vulkan kernels. KTT provides advanced autotuning features such as support for both dynamic (online) and offline tuning, and an ability to tune multiple kernels together with shared tuning parameters. Furthermore, it offers customization features that make integration into larger software suites possible. The framework handles all major steps required for autotuning implementation, including configuration space creation and exploration, kernel code execution and output validation. The public API is available natively in C++ and via bindings in Python.http://www.sciencedirect.com/science/article/pii/S235271102300081XAutotuningGPU optimizationCUDAOpenCLVulkan |
| spellingShingle | Filip Petrovič Jiří Filipovič Kernel Tuning Toolkit SoftwareX Autotuning GPU optimization CUDA OpenCL Vulkan |
| title | Kernel Tuning Toolkit |
| title_full | Kernel Tuning Toolkit |
| title_fullStr | Kernel Tuning Toolkit |
| title_full_unstemmed | Kernel Tuning Toolkit |
| title_short | Kernel Tuning Toolkit |
| title_sort | kernel tuning toolkit |
| topic | Autotuning GPU optimization CUDA OpenCL Vulkan |
| url | http://www.sciencedirect.com/science/article/pii/S235271102300081X |
| work_keys_str_mv | AT filippetrovic kerneltuningtoolkit AT jirifilipovic kerneltuningtoolkit |