You can use both Windows & *nix for deployment server.
Number of UF/deployment server depends upon polling interval. If you set polling interval of 30min, you can manage 1000+ UFs. Generally enterprises follows ITIL processes to push any new configuration on UF. Hence polling interval could be in hours as well.
A small deployment server (30 or fewer clients) can co-reside with a splunk instance which has other duties, such as a search head, indexer, or other splunk instance.
At moderate to large sizes (30-300), the deployment server should reside on its own splunk instance which does not have other duties.
Hope this can help you.
the given link is not working
Existing search head can work as DS as well depending upon number of UFs. DS cant be installed on indexers.
ok thank you.. What is the ideal time for polling by client to deployment server.?