With the rapid advancement of artificial intelligence and high-performance computing,heterogeneous computing platforms have evolved to encompass increasingly diverse architectures.While SYCL,an open standard for heter...With the rapid advancement of artificial intelligence and high-performance computing,heterogeneous computing platforms have evolved to encompass increasingly diverse architectures.While SYCL,an open standard for heterogeneous programming,has gained widespread adoption,its mainstream implementations(such as DPC++and AdaptiveCpp)primarily target SIMT-architecture devices like GPUs,presenting substantial challenges when adapting to specialized accelerators such as the Cambricon MLU,which employs a fundamentally different SIMD execution model.This cross-programming-model extension encounters two critical challenges:(1)bridging the programming abstraction gap between SIMT’s thread-level parallelism and SIMD’s data-level parallelism;and(2)harmonizing SYCL’s unified memory model with device-specific memory architectures.This paper proposes a novel cross-programming-model SYCL extension methodology to achieve full SYCL support for SIMD architectures,demonstrated through a comprehensive implementation for the Cambricon MLU platform.Our approach introduces MLU-specific vector programming interfaces while maintaining compatibility with the SYCL standard,enabling seamless integration of SIMD-based accelerators into the SYCL ecosystem.To validate our methodology,we integrated the extended SYCL-MLU implementation into PaddlePaddle’s CINN compiler,achieving a geometric mean performance improvement of 9.14%across representative neural networks,including ResNet,YOLOv3,and BERT.This research significantly broadens the application scope of SYCL in heterogeneous programming and provides a systematic methodology for extending SYCL to other SIMD-based hardware platforms.展开更多
Investigating and modeling fluid flow in fractured aquifers is a challenge. This study presents the results of a series of packer tests conducted in a fractured aquifer in Freiberg, Germany, where gneiss is the domina...Investigating and modeling fluid flow in fractured aquifers is a challenge. This study presents the results of a series of packer tests conducted in a fractured aquifer in Freiberg, Germany, where gneiss is the dominant rock type. Two methods were applied to acquire hydraulic properties from the packer tests: analytical and numerical modeling. MLU (Multi-Layer Unsteady state) for Windows is the analytical model that was applied. ANSYS-FLOTRAN was used to build a two-dimensional numerical model of the geometry of the layered aquifer. A reasonable match between experimental data and simulated data was achieved with the 2D numerical model while the solution from the analytical model revealed significant deviations with respect to direction.展开更多
基金supported by the Beijing Science and Technology Planning Project(Grant No.Z231100010323007).
文摘With the rapid advancement of artificial intelligence and high-performance computing,heterogeneous computing platforms have evolved to encompass increasingly diverse architectures.While SYCL,an open standard for heterogeneous programming,has gained widespread adoption,its mainstream implementations(such as DPC++and AdaptiveCpp)primarily target SIMT-architecture devices like GPUs,presenting substantial challenges when adapting to specialized accelerators such as the Cambricon MLU,which employs a fundamentally different SIMD execution model.This cross-programming-model extension encounters two critical challenges:(1)bridging the programming abstraction gap between SIMT’s thread-level parallelism and SIMD’s data-level parallelism;and(2)harmonizing SYCL’s unified memory model with device-specific memory architectures.This paper proposes a novel cross-programming-model SYCL extension methodology to achieve full SYCL support for SIMD architectures,demonstrated through a comprehensive implementation for the Cambricon MLU platform.Our approach introduces MLU-specific vector programming interfaces while maintaining compatibility with the SYCL standard,enabling seamless integration of SIMD-based accelerators into the SYCL ecosystem.To validate our methodology,we integrated the extended SYCL-MLU implementation into PaddlePaddle’s CINN compiler,achieving a geometric mean performance improvement of 9.14%across representative neural networks,including ResNet,YOLOv3,and BERT.This research significantly broadens the application scope of SYCL in heterogeneous programming and provides a systematic methodology for extending SYCL to other SIMD-based hardware platforms.
基金supported by the Department of Hydrogeology at TU Freiberg.
文摘Investigating and modeling fluid flow in fractured aquifers is a challenge. This study presents the results of a series of packer tests conducted in a fractured aquifer in Freiberg, Germany, where gneiss is the dominant rock type. Two methods were applied to acquire hydraulic properties from the packer tests: analytical and numerical modeling. MLU (Multi-Layer Unsteady state) for Windows is the analytical model that was applied. ANSYS-FLOTRAN was used to build a two-dimensional numerical model of the geometry of the layered aquifer. A reasonable match between experimental data and simulated data was achieved with the 2D numerical model while the solution from the analytical model revealed significant deviations with respect to direction.
