This is one of the most disappointing classes I ever took on Coursera. I am a Coursera believer and have taken so many classes here. But with all due respect, the instructor in week 2 and 3 was just mumbling to himself in front of the camera. I feel kind of disrespected. This course is not well prepared and should not be on Coursera.

The instructor in week 2 makes me want to kill myself. There are people who can teach, there are people who can't teach, and there is this guy who absolutely cannot be allowed to teach due to safety reasons.

Outdated (like what... 5y old?!)

A LOT of struggle with the old VM

Week1 is 1:1 the cloudera quickstart tutorial not even with additional comments

Liked the week 4&5 modules though but had to install spark locally

The course presents a general introduction to Hadoop world, but some explanations aren' t clear.

I appreciated links and references proposed to study other (ex. Python), and section quiz.

Are there problems with guides of Cloudera, but can be solved searching on the web (thank you Stackoverflow).

FInally, I didn't like sections presents by the Indian professor, because he "ate the words" and without underline text a lot of phrases were incomprehensible

With all due respect, I felt that the instructors tried their best to present a good introduction to Hadoop and its ecosystem. But the presentation fell short in many aspects. For example:

1. In the first 2 weeks of the course, I felt the instructors attempted to cover a lot of ground introducing all the parts of the Hadoop ecosystem and unfortunately none of the introduction was deep enough for me to remember.

2. Some of the lessons even attempted to talk in detail about the different variables and values needed to fine tune the Hadoop system. I am here to get an introduction to Hadoop, not fine tune it.

3. Outdated instructions getting pyspark to work for Cloudera VM.

4. Some questions in the quizzes at the end of each module were downright annoying and silly. They seemed to be testing the trivia of the Hadoop system rather than testing the big picture understanding.

5. The difference between spark and mapreduce is not explained fully. I was hoping that week 5 would give a thorough rundown of what spark is, how it works and why we need it, but not much of a ground was covered.

In retrospective, I feel that it would be best if the instructors redo this course properly addressing all these shortcomings.

This is a good introductory course. The content is good, but the presentation is not great. The presenters speak with a low, mumbling voice.

The programming exercises could also be improved: they are based on a contrived example (TV shows and view counts). The presentation of the exercise gives several clues on how to progress on the exercise, but in at least one case, that was not straightforward and there was a simpler solution.

Good enough for level 0 beginners, but it feels kind of scratches the surface. I'd say it takes you from level 0 to 0.8, not level 1 or 2.

The design of the final programming assignment was bad, not in terms of difficulty, but what it teaches you and expects from you. Seems like there is a problem with the right answer for it, but it's not being updated for 4 years. Also, the automatic grader doesn't give you any guide to get you correct your previous answer; you have to figure out why your submission is wrong with no clues. The worst part was that I had to correct the input file on my own, which should have been given rather than generated during one of the previous steps of the other assignments.

I took this course as part of "Big Data" specialization.

Unfortunately after taking break for half year the specialization changed and currently this course is not part of active "Big Data" specialization.

I cannot buy the complete specialization and cannot take single courses because specialization surprisingly can change so I prefer not to take more courses by this platform.

Thanks!

I hope they update the virtual machine provided and the scripts. Getting the VM to work in its current state requires familiarity with linux (in order to solve the permissions issues among other things).

As for the theory and the pace of the lectures. I really concerned myself to have learnt about these subjects.

To the lecturers and the staff, Thank you very much!

Well, the course is excellent for those from Hadoop or Apache Spark background, but as a newbie, please don't expect a nice detailed explanation from this course. Also, I was a newbie too and had to struggle more than usual to just pass the assignments especially programming assignments. But besides coding, architectural fundamentals are explained good enough to understand the basic working of Hadoop.

This is a very bad course. The VM image provided is very buggy and does not work. ( It took me 4 hours to make it to the second part of the first tutorial just to find out that the second part is broken due to miss-configured HUE server ). Professors do not respond to questions in forums as well. PLEASE do not waste your money/time on this course. The main issue is that other courses in the same specialization are very good but without passing this course, you can not earn specialization.

I was almost done with this course but I ended up finishing it. I have taken other courses on Coursera, and this one is the worst so far. You may get to know the basics of the Hadoop Platform but the lectures aren't much interesting and the lecturer reads the bullet points or a script. I was disappointed, frankly.

A bit more explanation along the slides would be really appreciated. Also it was a bit disappointing was the fact that the hands-on exercises were every word read from Cloudera, no addition/ additional information or clarification.

A walk-through of Cloudera's tutorials. Way overrated and over-hyped. Not worth the money.

Great course. I really liked the lectures. They were quite information-filled, so I really needed to pay attention. But I think I learned a lot and got a lot out of the course after knowing nothing about Hadoop. You do really need to know Python beforehand. Writing out Map and Reduce tasks and Spark jobs is hard enough with the syntax.

Great simple course to get started in the Hadoop zoo. The VM provided makes getting set up easy enough and is a great tool for assignments. I'd encourage the course creators to keep the VM up to date however, as it's beginning to accrue technical problems. Make sure you skim the forums for issues with the VM!

Super hands on introduction to key Hadoop components, such as Spark, Map Reduce, Hive, Pig, HBase, HDFS, YARN, Squoop and Flume.

I can't wait to the next course on the specialization.

I enrolled in this big data specialization with high hopes, I had hoped that I could gain a basic but very solid understanding of the different components of the Hadoop framework, with some meaningful assignments. After taking the course, I feel it's not really worth the time or money (50+dollars for a course). I have detailed complaints as follows:

1, The course lists way too many simple concepts, and also very repetitive. For example the Hadoop components introduction was not only too long, it was even repeated in the following course Big Data Analytics, and the previous course 1. The examples in the demonstration was also too simple and again repetitive. The word count case was used many times. Can you imagine a programming course teaching you to write "Hello word!" in 10 different ways? This is exactly it. Overall I feel the courses have not been well designed, but rather rushed through. Having multiple teachers to teach the same course was also a bad idea in my opinion, because the following lecturer seems to have not much awareness of the details that the previous lecturer did or did not talk about, this was obviously shown in several repeating(again) introduction of concepts.

2, Not impressed with the lecturer's presentation skills either. They seem to mostly just talk/recite in front of a monitor and camera, reading their presentations. I feel they just rush through the presentation in a very steady pace. They don't even do the in class examples themselves, just show a few line of codes and a snapshot of the result. You have to figure out all the hidden pieces.

3, The assignments. They are not well thought of. Some of those questions do not present anything meaningful. It's like testing if you slept through the presentation or not, even if you did you just need to go back to the presentation file and look for the answer.

4, Not a big fan of the choice of several software. The course already introduces you to so many key components of the Hadoop ecosystem, why introduce even more that are just a small player in the system? Serious, why KNIME? Why not just focus on Spark in the machine learning course?

Final thought,

The course does provide a good introduction to the Hadoop ecosystem, I also find the Cloudera VM to be a very good tool for practicing and learning, saving you a lot of trouble to install the whole system on a linux system. The course that I learned something really useful was mostly the classes on Map/Reduce and Spark, and some SQL like assignments in Pig/Hive. I hope that the lecturers can further refine and condense the course materials, and put more value into this specialization.

The quality of the course materials is sub-par. The five modules do not build up on one another. The first three spend a lot of time on the overall architecture of the Hadoop ecosystem, then the 5th module crams in Spark architecture and coding in one hurried module.

The quality of the videos is mediocre; the image of the instructor is often blocking the content on the slides so you can't read the whole slide, and the quality of the sound on some of the videos is of poor quality. Even the subtitles have obvious mistakes like "biplane" instead of "pipeline."

The instructions for the coding exercises are not well-organized, which make them harder to follow. The grader gives no feedback on an incorrect submission other than a generic "sorry, please get help from the forums" message.

Finally, there is almost no activity from TAs or other course staff on the forums. Many questions go unanswered.

This is all different from my previous experiences on Coursera, where the quality of the presentations and instruction and support from the course staff was top-notch.

I hope you can address these issues so this specialization can really offer quality content.

Awful! All type of content in the course are really bad: presentation, english level(grammar mistakes, style), material(lots of duplicate information, outdate information), tutorials (they are can't be run just of outdate instructions there)

The course starts to get weighed down by all of the material being shared, but just in the nick of time come the hands-on exercises. There are only a handful, and they come with pseudocode, so you aren't on an island, but they still add a ton of value because they put you in the driver's seat and get you thinking in terms of how you might solve a real-world problem in MapReduce or Spark.

Overall the course is very good. It helps you gain systematic knowledge on Hadoop and some very popular applications. But I have to say that the coding assignments are a little bit hard for those who don't have any programming experience. I would suggest add more hints in the assignments so that students do not need to put too much time researching about the basic syntax...

This course is very useful and provide technical insight about Hadoop and Spark framework architecture. This course teach you about what is big data and how to develop application using the Hadoop platform and I would like to recommend this course for those who want to become big data developer, this course is very helpful to start your career and understand the framework.

All the labs were really helpful in building the understanding of different applications. One area I felt could be helpful for labs is example of what output would look like at key intermediate steps. Writing up code stepwise and comparing results against some sample example would help as it would re-affirm the solution works as expected all the way through the exercise.

The course gave me an entry into Hadoop and Spark. I recommend revisiting the lectures and reading the forums if you get stuck. You will need a machine with enough RAM to run the Cloudera virtual box to do the programming assignments. I got stuck on a ubuntu 14.04 linux machine with only 8 gigs of ram but was able to run it on a iMac with 20 gigs of ram.