Introduction to High Performance Computing

2020 Stanford Summer Session “Introduction to High Performance Computing”

Instructed by Steve Jones
Course Description : Fundamentals of HPC, Parallel computer architectures, Cluster operating systems, and Resource management.

Tutorial hours : OpenHPC, Slurm workload manager, and Intel Parallel Studion.

Term project subject : A mathematical modeling and visualization of COVID-19 cases in South Korea.

Abstract: Kim, Soyoung et al. proposed the Modified SEIR Model for predicting the spread of COVID-19 in Korea. This is significant in terms of considering the behavior change of people like wearing masks or staying home. We implemented this model into python code using numerical methods and simulated the spread of COVID-19 based on confirmed cases on March 1st. We also visualized the simulation data and the actual data with the folium module to verify the model’s validity.

COVID-19_Visualization — COVID-19 Visualization

The model was based on a modified SEIR model. We utilized data from Kaggle to identify how many COVID cases existed in each province of Korea. The simulation was conducted on Python, and we used the folium module for visualization in the map.

With the simulation in HPC, we could estimate the number of COVID-19 cases in Korea accurately. Also, we could increase the computational performance with the Numba JIT compilation.