High - Performance Computing Engineer简历模板

During school, systematically studied professional courses such as computer architecture, algorithm design and analysis, and parallel computing. Participated in research projects in the school's high - performance computing laboratory, assisted the tutor in completing numerical simulation experiments based on cluster systems. By optimizing algorithms and resource allocation, the experimental efficiency was increased by 20%.

• Responsible for the daily operation and maintenance and optimization of the company's high - performance computing cluster, ensuring the 7×24 - hour stable operation of the cluster. The average cluster utilization rate was increased from 60% to 80%.
• Led the development of a parallel computing optimization scheme for molecular dynamics simulation. Adopted the MPI and OpenMP hybrid programming model, increasing the calculation speed by 3 times. This scheme has been applied in multiple scientific research projects of the company.
• Closely cooperated with the scientific research team, customized high - performance computing solutions according to project requirements, successfully supported the computing tasks of 5 major scientific research projects, and saved about 30% of the computing cost.

• Led a team (5 people) to undertake the CFD high - performance computing project in the aerospace field. Optimized the solver algorithm, reducing the complex flow field calculation time from 72 hours to 48 hours. The project results were highly recognized by customers.
• Designed and implemented a high - performance computing platform based on GPU acceleration. Realized the acceleration of specific computing modules through CUDA programming, increasing the overall computing performance by 5 times. This platform has served more than 10 projects inside and outside the company.
• Formulated the company's high - performance computing technical specifications and processes, organized internal technical training (cumulatively trained 200 person - times), and improved the overall technical level of the team.

• Project Background: A biomedicine company needed to perform molecular docking calculations on massive protein structure data, and the traditional calculation method was too time - consuming.
• Project Content: As a core member, designed a distributed computing architecture and used the Slurm job scheduling system to manage computing resources. Adopted a heterogeneous computing (CPU + GPU) mode and optimized the molecular docking algorithm.
• Project Results: The built high - performance computing platform reduced the computing task processing time from the original 1 month to 1 week, successfully docked more than 1000 protein - drug molecule pairs, and provided key data support for drug research and development.

• Project Background: The meteorological forecast center needed more accurate and rapid numerical weather forecast calculations.
• Project Content: Participated in optimizing the high - performance computing implementation of the meteorological model, analyzed the calculation bottleneck of the model, and adopted regional decomposition and load balancing strategies. Introduced a machine learning model to assist in the preprocessing of initial field data, reducing the amount of invalid calculations.
• Project Results: The calculation speed of the optimized meteorological model was increased by 40%, and the forecast accuracy was increased by 5%. It played an important role in the meteorological forecast during the 2020 flood season and successfully forecast the paths of 3 severe typhoons.