I developed these notes and exercises as part of a tutorial on how to use the Kao Group’s computing cluster. Although some of the details are specific to this specific cluster, much of the material could be useful for anyone getting started in computational physics, so I thought I would share it here. The materials are posted on github.com/adazi/bootCampEx and the best place to start is by reading README.md
Friday was my last day as a postdoc at NTU. My next step isn’t another postdoc or even a faculty position; instead, I’ll be learning about public policy as a AAAS Science & Technology Policy Fellow (STPF)! My placement is in the Department of Energy, where I’ll be working on the diplomatic and legal arrangements that support international scientific collaboration.
I was flipping through the fourth edition Landau and Binder’s excellent book on Monte Carlo for statistical physics and I came across this gem on p. 139:
We end this chapter by summarizing a few procedures which in our experience can be useful for reducing errors and making simulations studies more effective. These thoughts are quite general and widely applicable. While these ‘rules’ provide no ‘money-back’ guarantee that the results will be correct, they do provide a prudent guideline of steps to follow.
(1) In the very beginning, think.
What problem do you really want to solve and what method and strategy is best suited to the study. You may not always choose the best approach to begin with, but a little thought may reduce the number of false starts.
(2) In the beginning think small.
Work with small lattices and short runs. This is useful for obtaining rapid turnaround of results and for checking the correctness of a program. This also allows us to search rather rapidly through a wide range of parameter space to determine ranges with physically interesting behavior.
(3) Test the random number generator.
Find some limiting cases where accurate, or exact values of certain properties can be calculated, and compare your results of your algorithm with different random number sequences and/or different random number generators.
(4) Look at systematic variations with system size and run length.
Use a wide range of sizes and run lengths and then use scaling forms to analyze data.
(5) Calculate error bars.
Search for and estimate both statistical and systematic errors. This enables both you and other researchers to evaluate the correctness of the conclusions which are drawn from the data.
(6) Make a few very long runs.
Do this to ensure that there is not some hidden time scale which is much longer than anticipated.
The discovery of nuclear weapons might be the most consequential discovery that physicists will ever make. If you disagree, you will certainly agree with my hope that this discovery does not become any more important. I believe physicists have a special responsibility to both understand the legacy of nuclear weapons and help society to prevent them from ever being used again.
Last September, I visited the Hiroshima Peace Memorial Museum, a deeply moving testament to the horrifying consequences of war. While I was there, I purchased this book. It’s short and excellent telling of the human impact of the bombing. I highly recommend it, especially for my fellow physicists.
My Goodreads rating: 5 of 5 stars
“A short and beautiful book focusing on the human tragedy of people affected by the atomic bombing of Hiroshima and the lives they built in the aftermath.”
I always look forward to the APS Office of Government Affairs’ monthly Signal Boost video. This month’s update was full of great stuff!
I’m thrilled to announce that the Materials Modeling Stack Exchange forum is now in public beta. This means that anyone can browse without having to sign up for sign up for an account and the questions might start showing up in google search results. We’re still actively recruiting more physics-oriented contributors, so I encourage you to check it out.
There are already hundreds of questions and answers on the forum, here’s a couple great discussions you might want to join in on:
- What is a good programming language to learn for materials modeling?
- How to model phase transitions at critical regions in a magnetic system?
- How can very small lattices be sufficient for Quantum Monte Carlo simulations?
- Should I buy a CPU or a GPU for doing calculations?
- In Monte Carlo: does nonequilibrium imply stationary state?
- [Unanswered]: Discrepancy between numerical and transformed derivatives
- What are examples of materials that closely correspond to the Heisenberg model?
Scientists watch a lot of talks, and I’ve noticed a lot of people (including me) make the same handful of mistakes. Here are a few of my tips:
- Number your slides. Powerpoint, Keynote and Beamer all have options to add these automatically. Visible slide numbers make it easier for people to refer back to a specific slide if they have a question, especially at the end.
- Test your slides on a projector or low-resolution monitor. Computer monitor resolutions have steadily grown, but projectors technology seems stuck in 2004. This leads to a familiar trap: you make beautiful figure with graceful thin lines on your laptop, which are rendered totally invisible by the projector. Same goes for contrast, light colors like yellow are often invisible on projectors.
- Keep the text to a minimum. You want people listening to you speak, not reading your slides. Use slides for short bullet points and for showing off your figures.
- Even fewer equations. Unless you’re teaching a class, people are rarely going to be interested in following any mathematical derivations, and they’re hard to follow on a slide anyways. 1-2 equations per slide max. If people want to know more, they can always ask, which will probably lead to a more interesting discussion anyways.
- Finally, include your contact information on the final slide. It’s easy to space out at the start of a presentation and forget to jot down the presenter’s name. Make it easier for your audience by having your name and email on the last slide along with any relevant papers you want to promote.
Disclaimer: I want to be 100% clear that these tips are not a veiled reference to anyone in particular.
APS President Phil Bucksbaum recently wrote a letter with recommendations for how congress can protect science during COVID-19 and ensure a quick recovery afterwards. “The letter’s recommendations include: providing grantees full or partial cost extensions, ensuring the supplemental funding necessary to restart labs and experiments is provided, and substantially increasing REU funding for Summer 2021.”
APS is also organizing a letter-writing campaign to call Congress’s attention to this issue. They’re provided a easy-to-use tool where you can plug in your voting address, sign your letter (and add some of your own thoughts) and they will send it off to your congressperson and senators. It takes less than five minutes and it makes a huge difference. On narrow issues like this, you letter might be the only one your elected official receives!