Online Seminar: Serial Farming and Monte Carlo for SHARCNET

Speaker: Sergey Mashchenko, SHARCNET
In this talk, a specific research project (optical design of complex lenses) will be used as an example of advanced Monte-Carlo-type project setup in the SHARCNET environment. The following important ingredients of the setup will be covered in some detail: GSL (GNU Scientific LIbrary; specifically, the random number generation and multi-dimensional optimization parts); signal handling (both in code, and in bash scripts); bash scripting; the relevant aspects of the SHARNCET environment. The setup allows for an unlimited number of independent serial runs, ran at different times on different clusters, with the random number sequences never repeating.

No tickets available.