Coursera: Big Data Analysis with Scala and Spark
Date: 26/04/2017

By Yuri Kudryavcev

Work has been quite intense and very high-level recently, so I was missing the state of flow and the builders high that accompany coding. And I thought it's a good time to learn something new and since I've been reading heaps of good things about Apache Spark, a course on it sounded like a worthwhile pursuit. With that in mind, I enrolled in Big Data Analysis with Scala and Spark and thoroughly enjoyed the experience.

It turned out to be an outstanding course, Heather Miller did an excellent job fitting a lot about Spark architecture, processing and SparkSQL, an impressive variety of content during the short timeframe. I loved the coding assignments, they were all real-world based and fascinating to do.

Spark itself is breath-taking, I'm impressed by the performance, RDD architecture and the expressive power of Scala, it's something way better suited to a lot of data-oriented tasks than Map-Reduce.

On a side note, I enjoyed reading the Spark research papers, I found it a better architectural reference than any of the books I was reading. And it turns out even the sight of the 2 column layout with footnotes brings back all the -happy- PhD memories :)

There were a couple of things I didn't take into account when signing up:

  1. This course is the 4th course in 5-course Scala specialisation. I have never wrote anything in Scala before, and my last foray into functional programming was doing Scheme back at uni. Maybe it wasn't so smart to try to learn two new things at the same time. A good lesson in humility.
  2. More importantly, I completely underestimated my work commitments over the course period; I even ended up going overseas a couple of times (China and Singapore) during the course. Funny enough, those 8+hour flights are priceless for coding. It was a real challenge to meet the deadlines, and I was a week late in the end.

All in all great experience, I still have two books about Spark to finish up, and then I'll need to do something more 'real' with Spark. And I'll be back to the other courses in this specialisation once I get some breathing space. 

For more blogs from Yuri, please visit 

If you're looking to upskill and gain maximum value from your IBM solutions, Cornerstone provides training services with access to IBM courses for basic software training and/or fully customised training packages specific for your environment and users. Call us on 1300 841 048 or contact us online to learn more.