Mailing-List: contact commits-help@hama.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@hama.apache.org
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
From: Apache Wiki <wikidiffs@apache.org>
To: Apache Wiki <wikidiffs@apache.org>
Date: Thu, 24 Jan 2013 05:39:38 -0000
Message-ID: <20130124053938.55578.92065@eos.apache.org>
Subject: =?utf-8?q?=5BHama_Wiki=5D_Update_of_=22PageRank=22_by_edwardyoon?=
Auto-Submitted: auto-generated

Dear Wiki user,

You have subscribed to a wiki page or wiki category on "Hama Wiki" for chan=
ge notification.

The "PageRank" page has been changed by edwardyoon:
http://wiki.apache.org/hama/PageRank?action=3Ddiff&rev1=3D10&rev2=3D11

+ This document assume that you have already installed Hama cluster and you=
 have tested it using some examples.
+ =

  =3D=3D PageRank =3D=3D
  =

   * Uses the PageRank algorithm described in the Google Pregel paper
   * Introduces partitioning and collective communication
  =

- =3D=3D Usage =3D=3D
+ =3D=3D Run PageRank on Hama Cluster =3D=3D
+ =

+ First of all, generate a symmetric adjacency matrix using the gen command=
. =

  =

  {{{
- bin/hama jar ../hama-0.x.0-examples.jar pagerank <input path> <output pat=
h> [damping factor] [epsilon error] [tasks]
+   % bin/hama jar hama-examples-0.x.0.jar gen symmetric 100 10 randomgraph=
 2
  }}}
  =

- The default parameters for pagerank are:
+ This will create a graph with 100 nodes and 1K edges and store 2 partitio=
ns on HDFS as the sequence file. You can adjust partition and tasks numbers=
 to fit your cluster. Then, run PageRank using:
  =

  {{{
- 0.85 0.001
+   % bin/hama jar hama-examples-0.x.0.jar pagerank randomgraph pagerankres=
ult 4
  }}}
  =

- As you can see 0.85 is the damping factor, that is the probability which =
a user will "randomly" jump to other sides. See the [[http://en.wikipedia.o=
rg/wiki/PageRank#The_intentional_surfer_model|Random Surfer Model]].
+ =3D=3D Submit your own graph =3D=3D
  =

- 0.001 is the convergence error, the error will always be measured after a=
n iteration. It tells how much the pagerank of all sites has changed. If yo=
u are setting this to a lower factor, it will take more iterations. =

+ See [[WriteHamaGraphFile]]
  =

- =3D=3D Submit your own Web-graph =3D=3D
- =

- You can transform your graph as a adjacency list to fit into the input wh=
ich Hama is going to parse and calculate the Pagerank.
- =

- The file that Hama can successfully parse is a TextFile that has the foll=
owing layout:
- =

- {{{
- Site1\tSite2\tSite3
- Site2\tSite3
- Site3
- }}}
- =

- This piece of text will adjacent Site1 to Site2 and Site3, Site2 to Site3=
 and Site3 is a dangling node.
- As you can see a site is always on the leftmost side (we call it the key-=
site), and the outlinks are seperated by tabs (\t) as the following element=
s.
- =

- Make sure that every site's outlink can somewhere be found in the file as=
 a key-site. Otherwise it will result in weird NullPointerExceptions.
- =

- Then you can run pagerank on it with:
- =

- {{{
- bin/hama jar ../hama-0.x.0-examples.jar pagerank /tmp/input/input.txt /tm=
p/pagerank-output
- }}}
- =

- Note that based on what you have configured, the paths may be in HDFS or =
on local disk.
- =

- =3D=3D Output =3D=3D
- =

- The output is a double value that is between zero and 1.0. Where 1.0 is a=
 very "famous" site.
- =

- All pages' rank should sum up to 1.0, otherwise the algorithm is broken.
- =

- =

- =3D=3D Implementation =3D=3D
- =

- For detailed questions in terms of implementation have a look at my blog.
- It describes the algorithm and focuses on the main ideas showing implemen=
tation things. =

- It contains ancient code from before Hama 0.5 where we introduced the gra=
ph API.
- =

- http://codingwiththomas.blogspot.com/2011/04/pagerank-with-apache-hama.ht=
ml
-=20