Engineering topics
Zhang, Junchao
Publications and source records attributed to Zhang, Junchao.
Toward performance-portable PETSc for GPU-based exascale systems
The Portable Extensible Toolkit for Scientific computation (PETSc) library delivers scalable solvers for nonlinear time-dependent differential and algebraic equations and for numerical optimization. The PETSc design for performance portability addresses fundamental GPU accelerator challenges and stresses flexibility and extensibility by separating the programming model used by the application from that used by the library, and it enables application developers to use their preferred programming model, such as Kokkos, RAJA, SYCL, HIP, CUDA, or OpenCL, on upcoming exascale systems. Furthermore, a blueprint for using GPUs from PETSc-based codes is provided, and case studies emphasize the flexibility and high performance achieved on current GPU-based systems.
The PetscSF Scalable Communication Layer
PetscSF, the communication component of the Portable, Extensible Toolkit for Scientific Computation (PETSc), is designed to provide PETSc's communication infrastructure suitable for exascale computers that utilize GPUs and other accelerators. PetscSF provides a simple application programming interface (API) for managing common communication patterns in scientific computations by using a star-forest graph representation. PetscSF supports several implementations based on MPI and NVSHMEM, whose selection is based on the characteristics of the application or the target architecture. An efficient and portable model for network and intra-node communication is essential for implementing large-scale applications. The Message Passing Interface, which has been the de facto standard for distributed memory systems, has developed into a large complex API that does not yet provide high performance on the emerging heterogeneous CPU-GPU-based exascale systems. Here, we discuss the design of PetscSF, how it can overcome some difficulties of working directly with MPI on GPUs, and we demonstrate its performance, scalability, and novel features.
Evaluation of PETSc on a Heterogeneous Architecture, the OLCF Summit System: Part II - Basic Communication Performance
Nearest-neighbor communication is at the heart of many high-performance parallel computations. We report on the performance of such communication on the Oak Ridge Leadership Computing Facility system Summit in the context of the PETSc communication module. The analysis in this report includes basic Ping-Pong point-to-point communication and regular and irregular nearest-neighbor communication.
Effect of CeLa addition on the microstructures and mechanical properties of Al-Cu-Mn-Mg-Fe alloy
Development of high strength lithium battery shell alloy is highly desired for new energy automobile industry. The microstructures and mechanical properties of Al-Cu-Mn-Mg-Fe alloy with different CeLa additions were investigated through optical microscopy (OM), X-ray diffraction (XRD), scanning electron microscopy (SEM), transmission electron microscopy (TEM), Rietveld refinement and tensile testing. Experimental results indicate that Al{sub 8}Cu{sub 4}Ce and Al{sub 6}Cu{sub 6}La phases formed due to CeLa addition. Addition of 0.25 wt.% CeLa could promote the formation of denser precipitation of Al{sub 20}Cu{sub 2}Mn{sub 3} and Al{sub 6}(Mn, Fe) phases, which improved the mechanical properties of the alloy at room temperature. However, up to 0.50 wt.% CeLa addition could promote the formation of coarse Al{sub 8}Cu{sub 4}Ce phase, Al{sub 6}Cu{sub 6}La phase and Al{sub 6}(Mn, Fe) phase, which resulted in weakened mechanical properties. - Highlights: •Al-Cu-Mn-Mg-Fe alloys with different CeLa addition were fabricated through casting and rolling. •Al{sub 8}Cu{sub 4}Ce and Al{sub 6}Cu{sub 6}La phases formed after CeLa addition. •Addition of 0.25 wt.% CeLa promoted formation of denser precipitates of Al{sub 20}Cu{sub 2}Mn{sub 3} and Al{sub 6}(Mn, Fe). •Mechanical properties of the alloy was improved after 0.25 wt.% CeLa addition.