Parallel Particle-in-cell Performance Optimization: A Case Study of Electrospray Simulation

Open Access
Author:
Kodanganallur Narayanan, Ramachandran
Graduate Program:
Computer Science and Engineering
Degree:
Master of Science
Document Type:
Master Thesis
Date of Defense:
March 25, 2016
Committee Members:
  • Kamesh Madduri, Thesis Advisor
Keywords:
  • electrospray
  • multigrid
  • parallel
  • particle-in-cell
  • colloid thruster
Abstract:
The particle-in-cell (PIC) numerical technique is frequently used in physics and engineering simulations. In this work, we describe ES-PICBench, a new shared-memory parallel implementation of the PIC technique for electrospray simulations. Electrospray simulations are used in aerospace applications, and the goal of an electropray simulation is to understand behavior of an electrospray thruster or a colloid thruster. We discuss performance optimizations for various steps of a PIC-based electrospray simulation. One of the main steps in this simulation is solving the Poisson partial differential equation, and this step can be in turn converted to solving a system of linear equations. We develop a parallel implementation of the Multigrid method for this step. We demonstrate that ES-PICBench is significantly faster than other parallel PIC electrospray simulation implementations on Intel Xeon multicore platforms. Further, ES-PICBench can serve as a real-world scientific computing benchmark for analyzing parallel system performance.