Apache Spark with Scala By Example (Udemy.com)

When the instructor copies and pastes a command into the Spark Console, he doesn't wait very long sometimes before hitting enter, so if the output is more than one screenful, then the pasted command disappears in a flash, and you can't see what the command was unless you do a split-second screen capture to be able to study the command syntax.

This course gives you a flavor of Spark as well as same recipes but you already need to know the basics. Additionally, it is outdated, e.g. RDD's will be soon remove from Spark edition, and there's no reference on that.

In the middle of code demonstrations there are some errors on the console, it would be better to explain them more. Shared resource on troubleshooting guide of common problems on cluster deploy and other driver program runs would help more while learning in self-oriented way.

This course is a parade of quick examples in spark, none explained in much detail, none directly usable in real application, all shown in a hurry. Almost the same value could be achieved by simply providing the source code and a few inline comments for self study.

This course provides you with a jumpstart into Spark with Scala on AWS. The hands-on approach has the advantage of gaining confidence in e.g. the installation of a Spark cluster on AWS eliminating frustration typically associated with first-time attempts to install something. This course covers all these steps. Next, the Spark concept is well-explained with a good number of examples. Finally, the course addresses Spark Streaming and Machine Learning. The material received is definitely worth its money and I would recommend it as part of a curriculum in the context of Spark, Scala, and Machine Learning. Well-done!

I came to this course with reasonably high hopes. Online courses that help learning Spark with Scala are few and far between. In fact, Udemy themselves offer a 'Spark with Python' course. I didn't take that up because I don't know Python and I didn't want to be distracted. It is obvious to me that Todd has taken some effort to piece this course together. He is a hands-on technologist and this aspect becomes clearer as he takes us through the commands, shells, mistakes and recoveries. One very good feature of his curriculum is the treatment of RDD as a DB table (JDBC). This is uncommon. But on the whole, I am disappointed. Here's why: 1) Udemy should not dumb down such courses, for reasons to do with pricing or otherwise. This course is supposed to train us on Spark 'using' Scala. Therefore, at least a beginner's knowledge of Scala is a pre-requisite. 'sbt' (and if possible 'maven') is given. f I don't have that level of skills in Scala, I should not have been here. So, the declaration that even if I don't know Scala, I can appreciate this course, is giving a wrong picture and distorting the very objective of this course. Remember, we want to learn Spark here, not Scala. 2) The course begins at an exciting note when a publicly available dataset is downloaded and exploration begins. But, then, it fizzles out somewhere. A few more examples and exercises (the student need ideas about what can be found out: that's where the appeal of Spark lies) would have been much better. Todd may even go further and give us a few other links to explore in our own time, and come back and discuss with fellow students. Exercises (some with solutions, some for students to discuss amongst themselves and figure out) can add a lot more value to what is on the plate as of now. 3) There's no point in taking up the exercise of calculation of 'pi' as the piece to demonstrate writing and deploying a Scala application. That already exists in Spark's own site and in a multitude of other sites. As a learner, I relate to the act of writing the app, deploying it and then running it if it is 'my own application'. The dataset is already present; Todd could ask us to write a small program in Scala to derive one or two things meaningful, deploy that and see what happens. 4) If Todd asks us to run a Spark app using a largish dataset and then to turn over to the browser-based monitoring page, we can explore it with more relevant background. There is no point in showing a monitoring page where nothing is shown, running. Monitoring a Spark application is a very important aspect; a lot about how Spark expects us to run applications, can be understood from this page. In fact, Todd can make this exploration even more interesting by asking as to run an app (additionally) that fails. That way, monitoring a failed application is covered too. 5) Introduction to AWS is a very good aspect of this course (it helps underscore the point that running a Spark app in local mode is not where the world ends). However, I feel that this particular section begins excitingly but ends abruptly. Here again, Todd can take us through the journey a bit more diligently, asking us to deploy (and fail and learn). In the lecture, he inadvertently fails twice. To me, that is endearing because as a professional programmer I know that such hiccups happen. But, Todd can use that to teach us something. Udemy can actually help students with instructions of how to fire up an AWS instance for this purpose and more importantly, how exactly to close the instance down after the course is over. That ensures that a student incurs the cost of only what AWS resources s/he uses, to practice what the course teaches. 6) I also feel that in the SQL section, only one particular aspect should be brought in and then, at a little wider scale. In my opinion, as a Scala programmer, I want to spend more time on the transformations that are possible on a RDD. SQL is not really my focus here. So, going through Tables, JSONs and MySQL don't really add any much to my original objective of how to write Spark applications using Scala. 7) A bit gratuitous perhaps, but I think I should add this: a course like this should be specific: so, it can be offered in two sections one of which deals with Scala-based application writing and the other deals with interpreting data (using SQL, Graph or simple reports) that the foregoing course helps create. That way, there is a continuity; one relates better and perhaps, Udemy earns a bit more. Overall, I am disappointed. I would rather pay $80 or $100 to be confident about a non-trivial aspect of programming in the Big Data world, than I pay $25 for learning practically nothing. Udemy's other related course (Spark with Python) is much better is what I hear from my friends. There is indeed a great need of learning Spark using Scala. Udemy should exploit that while giving real value. I hope that Todd doesn't take this personally. He can transform this to a great course.

This course is very good for Hadoop developers who have some basic knowledge on Spark fundamentals. It was quick enough to grab some important concepts of RDD, DataFrames, CSV, JSON, MySQL JDBC, AWS, etc. I look forward to implement these at my work from day 2.

The instructor has really put some thought on his delivery and content. It covers the foundations of Spark in depth. I have done quite a few number of Spark courses, this course covers Spark Transformations, Actions and SparkSQL in great detail with easy examples to follow. It also covers how to connect to an RDBMS using JDBC from SparkSQL. Very rare examples on JDBC and SparkSQL. Please expand your course to add SparkR, MLIB and GraphX for it to become a killer Spark Training course. Thanks Todd, a job well done.

The course played out exactly as advertised. The modules were short and concise, containing valuable/pertinent information without getting bogged done in unneeded minutia. I come from a developer background, but a VERY different one than Spark. Even so, the instructions and exercises made sense and I found them valuable. I would definitely recommend this course to any one who wants to get up to speed quickly with Apache Spark.

This course is an excellent introduction to Apache Spark. The instructor is engaging and knowledgeable and offers a lot of helpful information in a clear and concise way. I even appreciated the little bits of humor here and there. I highly recommend this course.