Cluster Computing and MapReduce Lecture 2

Added: Aug 28, 2007

Duration: 52:5

Lecture 2: The MapReduce programming model. See http://code.google.com/edu/content/submissions/mapreduce-minilecture/listing.html for slides and other resources.

Channel: Howto

Tags: cluster computing distributed gfs hadoop mapreduce parallel

Rating: 4.32 (31 ratings) Views: 17704' favoriteCount='86 Comments: 9

yanglian Says:

Oct 25, 2007 - great! better than reading papers.

05cert Says:

Nov 15, 2007 - This is great, thank you google

jwalter2 Says:

Nov 18, 2007 - Thank you, Google. Google is my hero.

charlieok Says:

Nov 19, 2007 - Ironically, I'm having quite a bit of trouble watching this lecture series about reliable, highly available services. It's just not loading at all for me right now, but the rest of my internet access is fine.

nilbud Says:

Nov 22, 2007 - Input -> Process -> Output as we used to say in the olden days. 10 ? Hello 20 GOTO 10 RUN 0 0 0 0 0 0

thejiwon Says:

Dec 17, 2007 - What they're trying to do: exploit the grammatical characteristics of functional programming to allow concise expression of problems of inherent parallelism, while hiding the technicalities associated with the process of reducing them into parallel components.

thejiwon Says:

Dec 18, 2007 - I had the same problem - I paused it and left it that way until it was fully buffered, then played it and it plays well.

mrcoder Says:

Jan 2, 2008 - The lecturer's information flux density is overflowing my brain's buffers. He needs to dwell and linger on his points longer, and he needs to repeat them. This will give ample time for people to catch up with his ideas, assuming they are important enough that one should understand them. Basically what I'm saying is, his presentation is not casual enough. It's frustrating jibberish until I have replayed the video several times.

thecoolestjay Says:

Feb 8, 2008 - Either he's too smart, or he spent too much time preparing his lectures.