Warning: Undefined array key "DOKU_PREFS" in /usr/share/dokuwiki/inc/common.php on line 2082
cluster:121 [DokuWiki]

User Tools

Site Tools


cluster:121

Warning: Undefined array key 14 in /usr/share/dokuwiki/inc/html.php on line 1453

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Next revision
Previous revision
Last revision Both sides next revision
cluster:121 [2013/09/10 14:59]
hmeij created
cluster:121 [2013/09/16 11:06]
hmeij
Line 1: Line 1:
 \\ \\
 **[[cluster:0|Back]]** **[[cluster:0|Back]]**
 +
 +==== Hadoop Summary ====
 +
 +Our production Hadoop Cluster is based on [[http://www.cloudera.com/content/cloudera/en/home.html|Cloudera]]'s CD3U6 repository.  Here are some details :
 +
 +  * namenode (that is login node): whitetail.wesleyan.edu
 +    * whitetail also runs the Hadoop Scheduler and Health Monitor
 +      *  [[http://whitetail.wesleyan.edu:50070|Health Status]]
 +      *  [[http://whitetail.wesleyan.edu:50030|Job Tracker]]
 +    * ssh to it directly or from any of our other tails
 +  * resources: access to 600 GB of memory and 1.75 TB of Hadoop's Distributed File System (HDFS)
 +    * could be doubled in near future if needed
 +
 +  * HDFS is not backed up!
 +    * You must request a  writable work area /userdata/username
 +    * Be sure to down load your results to /home/username (that is the regular filesystem)
 +
 +  * Data to be shared (dictionaries, anagrams, etc) can be posted in /shareddata
 +    * request such items to be posted there
 +
 +  * Basic tools (request other tools to be installed)
 +    * shell scripting
 +    * python
 +    * perl (Hadoop::Streaming)
 +    * java (both Oracle in /usr/java and openJDK)
 +    * R+RHadoop (rmr2, rhdfs, rhbase)
 +    * Hbase (noSQL database)
 +      * [[http://whitetail.wesleyan.edu:60010|Master & Zookeepers]]
 +      * [[http://whitetail.wesleyan.edu:9095|Thrift server]]
 +    * MySQL 
 +      * request a database to be set up
 +
 +  * Note: the permissions are bit weird in HDFS but I think it is sorted out.
 +    * If this turns into a problem we'll let everybody run as user hdfs ...
 +  * Note: some http links will not work because they point to the private network
 +    * If you wish to view them launch firefox from whitetail ...
  
 Other useful pages Other useful pages
Line 7: Line 43:
   * [[cluster:115|Use Hadoop Cluster]]   * [[cluster:115|Use Hadoop Cluster]]
  
-==== Hadoop Summary ====+\\ 
 +**[[cluster:0|Back]]**
cluster/121.txt · Last modified: 2013/09/16 11:09 by hmeij