Re: [igraph] Igraph different performances

[Gábor Csárdi]

Marlos,

it is hard to say anything about this, since we do not even know which
igraph versions you use on the various machines, what other things are
running on the same machines, how big the graphs are, whether they fit
into the main memory, etc. Please provide more information. Thanks.

Btw. metrics 1, 2, 4 and 5 are hopeless for a very large graph. These
are quadratic algorithms.

Best Regards,
Gabor

On Fri, Oct 30, 2009 at 1:43 AM, Marlos C. Machado <address@hidden> wrote:
> Good Evening,
>
> I'm working with igraph-library and I have a weird situation. I need to
> calculate the following metrics for very large graphs:
>
> 1-igraph_average_path_length
> 2-igraph_diameter
> 3-igraph_transitivity_undirected
> 4-igraph_betweenness (average, so I do this all vertices)
> 5-igraph_closeness (average, so I do this all vertices)
> 6-igraph_density
> 7-igraph_clusters
> 8-igraph_maxdegree
>
> My problem is that it is taking too long for finishing. I'm running it in
> three different machines (M1 ... M3) and it is very slow on M1 and M3, but
> much faster at M2.
>
> M1: Intel(R) Core(TM)2 CPU          6320  @ 1.86GHz 2GB RAM
> M2: Intel(R) Core(TM)2 CPU          6320  @ 1.86GHz 2GB RAM
> M3: 3x Intel(R) Xeon(TM) CPU 3.00GHz 3GB RAM
>
> Did igraph suffer any change on the past months?  I think this faster
> machine has an older version of igraph.
>
> Thanks in advance.
>
> --
> Marlos Cholodovskis Machado <marlos at dcc.ufmg.br>
> Undergraduate student in Computer Science
> Department of Computer Science - DCC
> Federal University of Minas Gerais - UFMG
>
> _______________________________________________
> igraph-help mailing list
> address@hidden
> http://lists.nongnu.org/mailman/listinfo/igraph-help
>
>



-- 
Gabor Csardi <address@hidden>     UNIL DGM