cache_preload.txt

#!/bin/bash
#
# Preload a web site's cache
#
#---Doesn't quite work yet in the intended environment
#
#

site=$(echo ${PWD##*/})
tmp="preload"
log="log.txt"

echo "Crawling $site."

# Remove any prior downloaded files.
rm -rf $tmp

# Crawl the site
time wget --level=inf --directory-prefix=$tmp --output-file=$log -nv http://$site/

# When the crawl is done, the download files are removed.
rm -rf $tmp

echo
echo "Done.  A log of the crawl is in '$log'."

#End script

Just dump the above script into a file (preload.sh works nicely), put it into the site root folder (not web/content!) and set it up as a cron task to run every 24 hours. 

Credit for this to http://nadeausoftware.com/node/98#Howtopreloadthepagecache
Tweaks were made to this to make it more user-friendly for Cloud Sites customers.