Nandish Jayaram commented on MADLIB-1084:

This approach is using the Eigen Value based solution. I think the current 
MADlib implementation of PageRank uses the iterative approach instead.

You can certainly try to follow this approach, but I would suggest you check 
out what kind of matrix operations are already supported in MADlib. If you have 
to write corresponding matrix ops yourself, an iterative approach might be 
easier to implement for PPR. I am also not sure of the scalability of matrix 
operations in MADlib. For a billion node graph, the matrix will be very huge. 
If MADlib's matrix ops won't scale well, the overall runtime of PPR will take a 

[~fmcquillan], any thoughts on MADlib's matrix ops scalability numbers?

> Graph - Personalized PageRank
> -----------------------------
>                 Key: MADLIB-1084
>                 URL: https://issues.apache.org/jira/browse/MADLIB-1084
>             Project: Apache MADlib
>          Issue Type: New Feature
>          Components: Module: Graph
>            Reporter: Frank McQuillan
>            Assignee: Himanshu Pandey
>            Priority: Major
>             Fix For: v1.14
> Personalized PageRank which is a variant of regular PageRank.
> Please refer to  
> [http://madlib.apache.org/docs/latest/group__grp__pagerank.html] as a 
> starting point.
> Reference:
>  Neighborhood Formation and Anomaly Detection in Bipartite Graphs
>  [http://www.cs.cmu.edu/~deepay/mywww/papers/icdm05.pdf]

This message was sent by Atlassian JIRA

Reply via email to