Apache Hadoop - Petabytes and Terawatts

[aafreensheikh]

Couldn't find a better resource to get a quick overview of Hadoop and the plethora of related Apache projects.. Great presentation, great insight! Not to mention, great humor :)

[kumarchinnakali]

The best talk on HDFS and Map-Reduce. Thanks Jakob ! and looking more on the performance side of Hadoop. 

[bpriorb]

Excellent discussion on Hadoop. I am especially interested now in Giraph and will be checking it out. Thanks very much.

[edupulapatiajay]

It is very use full for HADOOP beginners, expecting more videos from You ... thanks

[oliverhu]

Great intro, still holds even it is already 7 year old.

[Sengup]

This is awesome. Great insight and information.

[karthik]

Great presentation. Very Informative!!!

[codingstrong]

Very educational and well explained presentation

[billypark]

love it! great information in such a short amount of time...

[AnandPritam]

Interesting viedo to understand what is going around hadoop

[pkumares]

Too good Presentation.. This guy is awesome

[govindgupta]

Very good presentation... Very Informative

[bhopalcoolboy]

It's so informative .. Amazing :)

[dylanhoggyt]

#33:50 whirr - automated cloud clusters on ec2, rackspace etc
#35:00 sqoop - relational data import
#35:55 mrunit - unit testing jobs
#36:20 mahout - machine learning libraries
#37:20 bigtop - interoperability
#37:35 crunch - MapReduce pipelines in Java and Scala
#40:00 Giraph - processing math on huge distribute graphs

[dylanhoggyt]

#21:30 pig - high-level mapreduce language
#23:10 hive - SQL like high-level mapreduce language
#26:10 hbase - realtime processing (based on google bigtable)
#27:40 accumulo - NSA fork of HBase
#28:40 avro - data serialisation
#30:30 zookeeper - low level coordination
#31:20 hcatalog - storage management and interoperability between all systems
#32:30 oozie - job scheduling
#33:20 flume - log and data aggregation

[Lanny]

The probability of multiple nodes going down at the same time is actually quite high. If you don't believe me, look up the Birthday Paradox (Problem) on Wikipedia.

[thorstenk]

very good presentation. What is the nice font, you are using?

[ravimshanbhag]

Any idea what was used to build this presentation? Doesn't look like it was Powerpoint on Windows.

[zantrua]

Why was hadoop seen as a new idea? It seems to me to be basically (foldl reducer (map maper data)), but in java.
Also, HDFS is just a bad DHT, from the looks of it.

[nkbuaa]

Let's do another version with Yarn.