NERSC is hosting an online webinar presented by William Arndt of NERSC and Geoffrey Lentner from Advanced Computing, Purdue University.
Background
Though high performance computing infrastructure is designed to maximize the performance of single large applications, it is just as capable of running high-throughput workloads that scale by running larger numbers of small compute tasks. Valuable scientific problems are well solved by employing this pattern of small and independent processes, such as those found in bioinformatics, high-energy physics, Monte Carlo methods, data processing, and more.
Topics covered
This training session will discuss and demonstrate multiple software tools, in order of increasing complexity and power, well-suited for managing high-throughput workloads: GNU Parallel, Snakemake, and Hypershell.
Additional attention will be given to designing effective arrangements of these tools, identifying performance bottlenecks, and best practices for improving the performance of throughput workloads.
Attendance details
The webinar is open to the general public but requires registration. It is geared towards Linux terminal users.
Please register now. A calendar invitation will be sent in advance of the training event.