Quantum chemistry and quantum simulation at scale : leveraging modern computing paradigms and hardware
Abstract/Contents
- Abstract
- Quantum chemistry has strong connections to important technologies and applications like quantum computing, machine learning, chemical or materials design, and drug discovery. At the heart of these connections is an increase in high-performance computing (HPC) power, largely driven by accelerators like GPUs. Improving how quantum chemistry methods are formulated and implemented to use current and emerging computing power is therefore crucial. In this work, I focus on two directions: scaling quantum chemistry methods to distributed, GPU-accelerated hardware to enable studying large chemical systems, and using tensor network methods to simulate and study quantum circuits for quantum computing. The fast evaluation of electronic structure is essential for studying the dynamics of molecular systems, for reaction discovery, or for generating high quality training data for machine learning models. I develop a multi-node, multi-GPU implementation of the electron-repulsion integrals—the computational bottleneck in many electronic structure methods—using the task-based parallel programming language Regent. This allows performing a calculation that treats the entire green fluorescent protein (GFP) molecule (3000+ atoms) quantum mechanically with high parallel efficiency. I go on to parallelize across molecular systems by implementing a distributed client-server ab initio exciton model, and present results using this model to study the excited state dynamics of the LH2 light-harvesting protein complex. Quantum computers are a promising emerging technology, but while real quantum devices remain error-prone and composed of small numbers of qubits, classical simulations of quantum circuits are important for studying the behavior of quantum algorithms. Here I present work using a classical simulator built on tree tensor network methods to study quantum circuits. I discuss the benefits and relevance of approximate methods like tensor networks which represent a quantum state in a compressed form, and discuss insights on Shor's algorithm for integer factorization.
Description
| Type of resource | text |
|---|---|
| Form | electronic resource; remote; computer; online resource |
| Extent | 1 online resource. |
| Place | California |
| Place | [Stanford, California] |
| Publisher | [Stanford University] |
| Copyright date | 2023; ©2023 |
| Publication date | 2023; 2023 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Author | Johnson, Katherine Grace |
|---|---|
| Degree supervisor | Martinez, Todd J. (Todd Joseph), 1968- |
| Thesis advisor | Martinez, Todd J. (Todd Joseph), 1968- |
| Thesis advisor | Aiken, Alexander |
| Thesis advisor | Markland, Thomas E |
| Degree committee member | Aiken, Alexander |
| Degree committee member | Markland, Thomas E |
| Associated with | Stanford University, School of Humanities and Sciences |
| Associated with | Stanford University, Department of Chemistry |
Subjects
| Genre | Theses |
|---|---|
| Genre | Text |
Bibliographic information
| Statement of responsibility | K. Grace Johnson. |
|---|---|
| Note | Submitted to the Department of Chemistry. |
| Thesis | Thesis Ph.D. Stanford University 2023. |
| Location | https://purl.stanford.edu/br930kh0109 |
Access conditions
- Copyright
- © 2023 by Katherine Grace Johnson
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...