=head1 SYNOPSIS
-
+ virt-similarity disk1 disk2 [disk3 ...]
=head1 DESCRIPTION
-Virt-similiarity is a tool for doing cluster analysis of groups of
+Virt-similarity is a tool for doing cluster analysis of groups of
virtual machines. It can automatically detect machines which have
been cloned from each other. It can produce a "cladogram" showing the
"family history" of each guest, or you can use it to create the most
efficient tree of backing files which will use the least disk space.
+Basic usage is simply to run virt-similarity with a list of disk
+images. It will read and analyze the disk images, then print a
+cladogram showing how they are related:
+
+ $ virt-similarity winxp-copy.img winxp-copy2.img \
+ f19rawhidex32.img f19rawhidex64.img archlinux20121201x64.img \
+ winxp.img winxp-copy2b.img
+
+ ---+------+------+------+------+------ winxp-copy2.img
+ | | | | |
+ | | | | +------ winxp-copy2b.img
+ | | | |
+ | | | +------+------ winxp.img
+ | | | |
+ | | | +------ winxp-copy.img
+ | | |
+ | | +------ f19rawhidex64.img
+ | |
+ | +------ f19rawhidex32.img
+ |
+ +------ archlinux20121201x64.img
+
+The first time you run virt-similarity, it has to read the disk images
+which takes some time. On subsequent runs, the information from each
+disk image is cached in a special form in a hidden file in your home
+directory called C<~/.similarity-cache.v1>. You can simply delete
+this cache file at any time if you want, especially if you want to
+force virt-similarity to re-read the disk images.
+=head1 FILES
+=over 4
+=item C<~/.similarity-cache.v1>
-=head1 FILES
-
+The cache file where virt-similarity saves information about each disk
+image so that it doesn't have to re-read them each time it is run.
+You can delete this file at any time.
+=back
=head1 ENVIRONMENT VARIABLES
-
+Virt-similarity uses libguestfs, and is affected by various libguestfs
+environment variables. See L<guestfs(3)/ENVIRONMENT VARIABLES>.
=head1 SEE ALSO
-
+L<guestfs(3)>, L<http://virt-tools.org/>
=head1 AUTHOR