[Rsiena-help] RSiena: Error using multiple processors

[Tom Snijders]

Dear Philippe,

The error seems not to be specific for the multiple processors. The main error message seems to be "Unlikely to terminate this epoch:  more than 1000000 steps". For this error, see the section in the RSiena manual on error messages; it is listed there.

I hope this and what is listed there in the manual helps - else, let us know!

Best wishes,
Tom

================================================================
Tom A.B. Snijders
Professor of Statistics in the Social Sciences
Department of Politics and Department of Statistics
Nuffield College
University of Oxford
tel. +44-01865-278599

From: rsiena-help-bounces at lists.r-forge.r-project.org [mailto:rsiena-help-bounces at lists.r-forge.r-project.org] On Behalf Of Philippe Sulger
Sent: 14 September 2012 10:13
To: rsiena-help at lists.r-forge.r-project.org
Subject: [Rsiena-help] RSiena: Error using multiple processors

Dear All (I've sent this already with another email-account, not subscribed to the list)

I am running a siena07 analysis on 102 networks. I do this via a loop and I am using a server with 16 cores. The command I enter is

for (i in c(1:102))
    {ans.basicI[[i]] <- siena07(basic.modelI[[i]], data = mydata.basic.model[[i]], effects = myeff.basic.modelI[[i]], batch = T, nbrNodes = 14, useCluster = T);
    sink(paste("basic.modelI.",i,".txt"));
    show(ans.basicI[[i]])
    sink()}

As you note, I decided for the time being to use only 14 cores. This also gives me some txt-files with basic estimates. Also if I exclude the sink()-txt-sink() part I receive the an error that looks as follows:

Start phase 2.1
Phase 2 Subphase 1 Iteration 1 Progress: 4%
Phase 2 Subphase 1 Iteration 2 Progress: 4%
Error in checkForRemoteErrors(lapply(cl, recvResult)) :
  14 nodes produced errors; first error: Unlikely to terminate this epoch:  more than 1000000 steps
In addition: There were 16 warnings (use warnings() to see them)
> warnings()
Warning messages:
1: closing unused connection 17 (<-localhost:10187)
2: closing unused connection 16 (<-localhost:10187)
3: closing unused connection 15 (<-localhost:10187)
4: closing unused connection 14 (<-localhost:10187)
5: closing unused connection 13 (<-localhost:10187)
6: closing unused connection 12 (<-localhost:10187)
7: closing unused connection 11 (<-localhost:10187)
8: closing unused connection 10 (<-localhost:10187)
9: closing unused connection 9 (<-localhost:10187)
10: closing unused connection 8 (<-localhost:10187)
11: closing unused connection 7 (<-localhost:10187)
12: closing unused connection 6 (<-localhost:10187)
13: closing unused connection 5 (<-localhost:10187)
14: closing unused connection 4 (<-localhost:10187)
15: In rbind(`1` = 2691L, `2` = integer(0), `3` = integer(0),  ... :
  number of columns of result is not a multiple of vector length (arg 25)
16: In rbind(`1` = integer(0), `2` = integer(0), `3` = integer(0),  ... :
  number of columns of result is not a multiple of vector length (arg 37)

The loop works fine up to i=16.

Excerting the code on my Mac works fine...Do you have an idea what could cause the problem?

Thank you for your help.

Best,
Philippe

________________________________
Nuffield College is a Registered Charity No. 1137506. Registered Office: Nuffield College, New Road, Oxford, OX1 1NF
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/rsiena-help/attachments/20120914/8531958a/attachment.html>