This code has been developed in the scope of the following work:
Sandeep Suresh Cranganore, Vincenzo De Maio, Ivona Brandic, Tu Mai Anh Do, Ewa Deelman: Molecular Dynamics Workflow Decomposition for Hybrid Classic/Quantum Systems. e-Science 2022: 346-356
Please cite accordingly: https://dblp.org/rec/conf/eScience/CranganoreMBDD22.html?view=bibtex
This repo contains code and notebooks about Quantum Computing experiments.
We acknowledge the use of IBM Quantum services for this work. The views expressed are those of the authors, and do not reflect the official policy or position of IBM or the IBM Quantum team. IBM Quantum
Machines used in this work:
| ID | VERSION | PROCESSOR |
|---|---|---|
| ibmq_manila | 1.0.29 | Falcon r5.11L |
| ibmq_santiago | 1.4.1 | Falcon r4L |
- quantumMD A preliminary study of applying quantum computing to scientific computation
- quantum-pqc-comparison An example of performance of different PQC for variational algorithms
Filename format: ALGORITHM_BACKEND_QUBITS_[RT-NRMSE|SUMMARY]
- rt-nrmse: summary of the experiment with average runtime and normalized root mean square error between the value obtained on classic architecture and the value obtained with specific PQC;
| PQC | AVG-RUNTIME | NRMSE |
|---|---|---|
| PQC name | Average RT for circuit | NRMSE for circuit |
- rawdata: data of each execution of VQE over which rt-nrmse are calculated.
| PQC0-RT | PQC0-EIG | ... | PQCn-RT | PQCn-EIG | Classic-EIG |
|---|---|---|---|---|---|
| RT matrix #1 using PQC0 | EIG matrix #1 using PQC0 | ... | RT matrix #1 using PQCn | EIG matrix #1 using PQCn | EIG matrix #1 classic |
| ....... | ........ | ... | ....... | ........ | ........... |
| RT matrix #m using PQC0 | EIG matrix #m using PQC0 | ... | RT matrix #m using PQCn | EIG matrix #m using PQCn | EIG matrix #m classic |
Note: Simulator assumes full qubit connectivity, therefore performance is always better than real backend for low number of qubits.