Hadoop Platform and Application Framework

I actually enjoyed the course, but -1 star for being a little outdated and -1 star for too much content and few practical exercises in the central parts. The structure is well thought though, and the PySpark lessons very entertaining. I would recommend this course for those interested in getting an historical perspective but warning them that MapReduce and Spark RDDs have been superseded with recent developments.

Not enough exercisesToo general

Good concepts but poorly organised exercises!

Course provides basics of Hadoop, MapReduce and Spark. In my opinion program assignments should be more formalised with no dependancy on programming language at all (plain result submission) OR with strict dependency (code submission).

The course overall is good; however, week#2 for example was pretty bad. It felt like a very theoretical style of teaching a course which could have been made hands on. Week #2, which i am in process of completing is SO MUCH theory. Tez, YARN, Spark... difference b/w hadoop and hadoop#2.. Do you want us to cram the differences? Show us how to build something, show us how to set-up a JOB... be practical.

The entire course is pretty good. But the final assignment with the N to N match with Program as the key should really caught attention of the mentor. With N to N match, the only "right" result is to right the very similar Pyspark code just like that write by who was writing the code to generate the answer. However, the N to N match, there is no true answer at all. Therefore, any code other than what the writer wrote should be given partial credit. All in all, a pretty good course, but assignment 15 definitely need to be rewritten to get rid of the N to N match.

Thanks for professors' work.

Good solid course with a lot of emphasis on hands-on! It would be great if the teachers (particular in the Spark/pySpark section) could provide setup guidance for iPython Notebook as this will save the students a lot of time in coding/re-coding various examples as well as having a complete and easy to overview trace of the various Python related exercises.

Challenging assignments made the course special. Its less theory and more practice and that's exactly what I'm looking for.

I found this course a lot better than the previous one.More concrete and with interesting material quite applicable on work use cases.I think there are some points of attention to be addressed in next versions of this course.a) Quiz sometime are very hard to be solved because there is no explanation in the course. But quiz are very useful (as the community) to really learn the material.b) some lessons are hard to follow. Probably better to put at the beginning of lessons a mandatory activity to read some material on web (I found very useful links in discussion community, after them I was able to follow the course, but not before)In any case I consider this very useful and I thanks all of you for the effort you put itStefano Priola

This is a good course for anyone without major experience with Hadoop and/or Spark. Covers high level concepts and architecture, and basic tools of each. In this first iteration of the course, there are several typos in the assignments but fellow students have quickly provided corrections that, I'm sure, will be incorporated into subsequent offerings.If you want to learn, I would recommend at least trying this course.

Good foundation for distributed file systems. Material dated? Presenters very capable, a little dry.