Hosting a cutting-edge Data Science learning experience with SWITCHengines: a success story from the University of St. Gallen.
Data is claimed to be the new oil in today’s world; but few have the necessary skills to create value out of this resource. To equip future professionals with these vital skills, professors Juan-Pablo Ortega and Johannes Binswanger of the University St. Gallen, designed a course to teach the fundamentals of Data Science. The Certificate course Data Science Fundamentals (DSF) is open to all undergraduate students eager to learn the basics of data handling, machine learning and statistical analysis.
During a course, 60 students learn to handle large volumes of data and how to code algorithms. That requires a lot of data to be set up, stored and processed. "We started the DSF program in 2017 and quickly realised that the technical requirements for this course are a challenge," remembers Juan-Pablo Ortega, Prof. PhD. Faculty of Mathematics and Statistics, University of St. Gallen.
«SWITCHengines enables location-independent teaching of large groups. Students can work on assignments at home and computational power and resources of machines can be adjusted according to the topic discussed in class.»Anastasija Tetereva, PhD, PostDoc for Consultancy in Empirical Research Methods at the Swiss Institute for Empirical Economic Research, University St. Gallen.
Apart from the computational power, this type of course requires a completely flexible working environment for students - flexible in terms of location, software and working tools. "We struggled with the technical solutions we had at first, and started looking for something different. With SWITCHengines we now have a course environment that works out-of-the-box," says Juan-Pablo Ortega.
The configuration process is simple; one individual engine is set up by installing all the required software and data distribution system. Copies for all students can be created with only one click. This image can then be reused to set up future courses.
With this set up, no software needs to be installed locally on laptops. Students can work on assignments at home with low risk of technical difficulties. Moreover, computational power and resources of machines can be adjusted according to the topic discussed in class. Each student is provided with an individual engine. This minimises the downtime risk in comparison to the set-up when many users are connected to one server.
Further aspects that convinced the University of St. Gallen to work with SWITCHengines were cost transparency and the support SWITCH provides, explains Anastasija Tetereva."Being part of the academic landscape, SWITCH understood our needs and helped us to find the best individual solution for our Data Science course."