This is the official repository for IPRO 497 - Antimatter Interferometer as of Fall 2020. All code needed to run the Optical Simulation can be found here. Any students in this course should first fork this repository and update any time specific information pertaining to when they took the course. For more information on contributing, visit the contributing section of this document. For any other information regarding the course, please contact Dr. Daniel Kaplan or Dr. Derrick Mancini.
The purpose of this IPRO is to experimentally answer if matter and antimatter repel each other through gravity. The method we use to answer this is to measure the effects of gravity on antimatter using muonium and seeing if antimatter responds to gravity in a measurable manner. The equipment to undergo this expirement is expensive and time consuming to make. A simulation of this experiment would prove useful because it is significantly easier to test prototypes or other changes without having to spend time or money on hardware that may or may not work.
You will need a computer with an NVIDIA GPU that supports CUDA.You will also need to use Linux. WSL may work. Virtualization will likely not work unless it offers GPU passthrough. We used a system with the following configuration:
(2x) Intel(R) Xeon(R) CPU E5-2630 v2 @ 2.60GHz
(5x) Micron 4GB MT9JSF51272PZ-1G9E2
(1x) TITAN Xp (1x) Tesla K40c
Install Conda with Python 3.6 or greater.
Install the NVIDIA CUDA Toolkit.
Once installed, you should edit /options.sh to update the path of the CUDA binary folder. By default, we use /usr/local/cuda-9.1/bin/ because we have multiple versions and want to use that specific version of NVIDIA CUDA.
Run conda deactivate to deactivate the base environment (if applicable).
Run conda env create -f environment.yml in this directory to create a Conda environment with all dependencies specified in environment.yml.
The environment is called agis.
Run conda activate agis to use the virtual environment configured with Conda for this project.
Run bash scripts/install_extra_conda_dependencies.sh to install dependencies we cannot specify in the environment.yml file.
Then run python -m optical_simulation.run_simulation to run the slit simulation.
You can also run python -m optical_simulation.demo.runMultiCoreTest to run the multi-core test.
There is a shell script called RunMultipleTests.sh that will run multiple simulations with slightly different parameters.
There are also profiling scripts called profile-*.sh.
Add the package to environment.yml.
Run conda env update -f environment.yml in this directory to update Conda's environment from this file.
One of the tasks from Fall 2020 was to make the whole simulation 3-dimensional. This was one of many tasks we had planned that semester so unfortunetly, there was not enough time left to finish. This task is a long-term goal and other groups should continue where we left off if they are willing to make it the theme for their entire semester. There can be many benefits to completing this task such as incorporating strut changes and exploring more up to date computing hardware.
Previously to Fall 2020, there was no default way to keep track of semester progress that we were aware of besides GitHub. To reduce confusion and ease the transition process between semester groups, please follow this section to make the appropriate changes that will signify which repo is the current version and to let future students know what each section of code does. This is crucial since a lot of time can be wasted trying to figure out what other people did.
- Fork the previous semester's repository.
- Update the Introduction section of the README.md file to reflect your current semester. Specifically, change the semester description in bold.
- Add as much documentation as possible to your semester's Google Drive folder. Write everything clearly as if a complete stranger will need to read it one day.
- Save all test results locally or somewhere outside of the GitHub repository commit history. Only save successful test results that have been used to show improvements in midterm or final presentations. Discard any results that were generated from faulty code or prototypes.
- COMMENT YOUR CODE.
The Background folder in Google Drive
Measuring Antimatter Gravity with Muonium
The Double-Slit Experiment
What is an Interferometer?
Muonium Wikipedia Article
Anaconda Cheat Sheet
CUDA Documentation