Ticket #96 (closed defect: fixed)

Opened 1 year ago

Last modified 3 months ago

fortran algorithms for directed networks not working for large network

Reported by: bherr@IU.EDU Assigned to: rduhon
Priority: low Milestone:
Component: Algorithm Version: 0.8.x
Severity: minor Keywords:
Cc:

Description

Every fortran algorithm in the Analysis/Unweighted & Directed menu all expect a directed graph, but always output that the program should be applied on undirected networks (some actually say directed networks, even if i'm feeding it a directed network). However, if I make it an undirected network (by changing the section from DirectedEdges? to UndirectedEdges?) it still fails. Basically these algorithms are completely broken. Attaching the nwb network file I used. It is a large file, but still passes the validator tests. I have gotten smaller directed networks to work, so i'm not sure what is going on.

Change History

09/25/07 11:40:05 changed by bherr@IU.EDU

  • summary changed from fortran algorithms for directed networks not working for large network to added network file location.

Tried to upload the network, but its too big. It can be downloaded from http://ella.slis.indiana.edu/~bherr/tmp/enwiki-20070802-network.nwb.gz

09/25/07 11:40:29 changed by bherr@IU.EDU

  • summary changed from added network file location to * fortran algorithms for directed networks not working for large network to added network file location..

09/25/07 11:41:02 changed by bherr@IU.EDU

  • summary changed from * fortran algorithms for directed networks not working for large network to added network file location. to fortran algorithms for directed networks not working for large network to added network file location..

ugh

09/25/07 11:41:36 changed by bherr@IU.EDU

  • summary changed from fortran algorithms for directed networks not working for large network to added network file location. to fortran algorithms for directed networks not working for large network.

09/25/07 12:01:12 changed by rduhon@IU.EDU

How large is large?

09/25/07 13:17:41 changed by rduhon@IU.EDU

  • priority changed from highest to low.
  • severity changed from critical to minor.

I verified that at least some of these work at least up through algorithms with two million nodes and eighteen million edges, approximately the same size as the wikipedia one that was tried. This might be a machine resource limitation, though my machine only has two gigs of RAM, but I think its working well enough. The problem might also be due to some particular characteristic of the network . . . actually, that makes me think.

10/04/07 17:41:46 changed by huangb@IU.EDU

  • version set to 0.8.x.
  • milestone set to 0.8.

08/14/08 15:50:18 changed by mwlinnem@IU.EDU

  • status changed from new to closed.
  • resolution set to fixed.

I just tested this, and it seems to be fixed, so I'm closing the ticket.