But let’s not forget that it was Ada Lovelace who kicked off the computer era in 1843 when she outlined a sequence of operations for solving mathematical problems with Charles Babbage’s Analytical Engine. Or that up through the 1960s, women actually were the computers and the primary programmers.
Times have changed, but women’s contributions to computing haven’t. So to correct some mistaken ideas, here are five cool things women are doing with high-performance computing.
Speeding up our understanding of the Universe
The Dark Energy Spectroscopic Instrument (DESI) survey will make the largest, most-detailed 3D map of the Universe ever created and help scientists better understand dark energy. Every night for 5 years, DESI will take images of the night sky that will be used to construct a 3D map spanning the nearby universe to 11 billion light years.
But in order for that map to be made, images from the telescope must be processed by the Cori supercomputer. Laurie Stephey, a postdoctoral fellow at Lawrence Berkeley National Lab (LBNL) is optimizing data processing for the DESI experiment so that results can be returned to researchers overnight in order to plan their next night of observation.
Developing fusion as a renewable energy source
Plasma is the fourth state of matter, made up of energetic, charged particles. Fusion happens when two light elements, like hydrogen, fuse together to form a heavier element, such as helium, and give off a lot of energy. This process happens naturally in stars like our sun, but scientists are working to recreate this in a lab.
Tess Bernard, a graduate student at the University of Texas at Austin is developing computer simulations to model the physics of plasmas in order to help design successful fusion experiments. Says Bernard, “If we can successfully harness fusion energy on earth, we can provide a clean, renewable source of energy for the world.”
Dealing with big data
Modern scientific computing addresses a wide variety of real-world problems, from developing efficient fuels to predicting extreme weather. But these applications produce immense volumes of data which are cumbersome to store, manage, and explore.
Which is why Margaret Lawson, a PhD student at the University of Illinois at Urbana-Champaign and Sandia National Laboratories is creating a system that allows scientists working with massive amounts of data to tag and search specific data. This makes it easier for scientists to make discoveries since the most interesting data is highlighted for further analysis.
Preparing for exascale
Exascale computing will represent a 50- to 100-fold increase in speed over today’s supercomputers and promises significant breakthroughs in many areas. But to reach these speeds, exascale machines will be massively parallel, and applications must be able to perform on a wide variety of architectures.
Abigail Hsu, a PhD student at Stony Brook University, is investigating how different approaches to parallel optimization impact the performance portability of unstructured mesh Fortran codes. She hopes this will encourage the development of Fortran applications for exascale architectures.
Computers make mistakes. And sometimes those failures have serious consequences. Like during the Gulf War, when an American missile failed to intercept an incoming Iraqi Scud. The Scud struck a barrack, killing 28 soldiers and injuring a hundred others. A report attributed this to computer arithmetic error--specifically a small error of 0.34 seconds in the system's internal clock.
Harshitha Menon, a computer scientist at Lawrence Livermore National Laboratory (LLNL) is developing a method to understand the impact of arithmetic errors in computing. Her tool identifies vulnerable regions of code to ensure that simulations give correct results.
Says Menon, “We need to understand the impact of these errors on our computer programs because scientists and policy makers rely on their results to make accurate predictions that can have lasting impact.”
More women, more science
So that covers astronomy, physics, computer science, and math. And they say women don’t like science. We say that’s a pretty unscientific conclusion.