Github user ankurdave commented on the pull request:
https://github.com/apache/spark/pull/2100#issuecomment-53311125
To answer your concern, the `runUntilConvergence` version of PageRank uses
delta messages, where the msgSum operates on the delta graph where the rank of
each vertex is the change in PageRank from one iteration to the next. You can
see that on lines 144 and 149: the `sendMessage` function uses
`edge.srcAttr._2` as the source rank, which was set to `newPR - oldPR`.
As a result, we have to add in `oldPR` at each step to obtain the PageRank
of the input graph rather than the delta graph.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
