Register to watch in person in the Kavli Auditorium, or watch the lecture live on our YouTube page. 

For a decade, SLAC has been using its X-ray laser, the Linac Coherent Light Source, to explore the properties of matter at the atomic level. One example, which is the subject of my own research, is the measurement of atomic defects in semiconductors that can be used in transistors. At extremely low temperatures, these materials undergo dramatic changes in structure that make them ideal materials for quantum sensors. Using intense X-rays, we can visualize the rearrangements of electrons and atoms that occur in these transitions. Already, these experiments generate massive amounts of data, on the scale of Terabytes per hour. Recently, we commissioned a second-generation X-ray laser, the LCLS-II, that will produce enormously more X-rays and significantly more data, up to Terabytes per second. How can we harness this flood of information? Our response has been to construct a new parallel computing platform called "cuPyNumeric" which enables automatic, massive data analytics on many graphics processing units (GPUs). This approach allows us not only to analyze the data much faster but also to understand it in real time. It enables us to feed back to the X-ray source, constantly steering the experiment toward optimum performance and speeding up scientific discovery. With this new capability, we can quickly turn novel layered materials imagined by scientists into well-characterized components ready for application.