Re: [Boost-users] Help on creating a custom InputIterator for boost graph adjacency_list()

Subject: Re: [Boost-users] Help on creating a custom InputIterator for boost graph adjacency_list()
From: Trevor Harmon (Trevor.W.Harmon_at_[hidden])
Date: 2010-05-20 15:32:51

On May 19, 2010, at 12:19 AM, Cedric Laczny wrote:

> I have a graph with quite some information stored in its vertices  
> and edges.
> Now I would like to perform some analyses on the general overall  
> topology of the graph. Therefore I don't need this "big graph" but  
> only a graph that has the same number of nodes and matching edges to  
> the "big graph".

I am wondering if this is a case of premature optimization. Are you  
sure that removing these many properties attached to the vertices and  
edges would actually speed up the analysis? I would expect there to be  
little if any difference. Also, it is possible that the task of  
creating the duplicate graph would offset any speed gains there might  
be in the subsequent analysis.

Trevor