Artificial intelligence (AI), data analytics and machine learning are fields in computer sciences with presence in many other research fields and supercomputing. With the goal of bringing AI methods to computer sciences students, also professionals and researchers of any field dealing with predictive analytics, the Barcelona Supercomputing Center (BSC) has created an on-line course on AI in High Performance Computing environments. This course is composed by video-tutorials, a brief theoretical introduction of concepts around Data Centers and Machine Learning, also examples and exercises using the data processing platforms Apache Spark and Intel BigDL.
In this on-line course, Josep Lluís Berral, creator of the course and researcher at the Data Centric Computing group, introduces basic concepts of parallelism, performance and virtualization, also machine learning, oriented to students and researchers either familiar or novel with those concepts.
The course is divided in three fundamental sections: performance in data centers, machine learning, and technologies for data processing. Each part provides a set of video-tutorials including theory and examples, with the corresponding slides, practical examples and exercises. In first and second parts, the tutorials present the fundamentals for supercomputing and artificial intelligence. In the third part, the distributed data processing and Deep Learning platforms Spark and BigDL are presented, with practical demonstrations. In total, Berral with the help of Francisco Javier Jurado, prepared more than five hours of video-tutorials, around the three chapters, along examples and exercises, and additionally a set of scripts to make easier to deploy the proposed environments.
“We have prepared this course thinking in those researchers working around us at the Barcelona Supercomputing Center and in partner research groups, that being from fields like Mathematics, Biology, Genomics or Earth sciences, have the need to use big computers to process large amounts of experiments, but never had the occasion to use frameworks that definitely can improve their daily work at the lab or office”, says Josep Lluís Berral.
Berral received in 2019 a sponsorship from the Universities Outreach department at Intel Corp., as part of the Academic Educational Mindshare Initiative for AI, to develop and publish the course.
The chosen technologies for this on-line course are known scripting environments like Python, virtualization tools like VirtualBox, the distributed data processing framework Apache Spark, and the libraries for Distribtued Deep Learning Intel BigDl. It is expected that the contents and proposed technologies will be updated, as the shown technologies evolve and the users send their feedback.
