Sedo, Afternic, BuyDomains, or here. Out of those choices, Sedo is probably your best bet because people searching for a particular keyword that won't be available in other gTLD's might notice .pro and think it sounds OK compared to some less striking ccTLD's. Even then, they willl have to be professionally qualified which rules out 90% of buyers.
Realistically, only top keywords that appeal to a wide range of developers, a tiny proportion of which will be professionally qualified and open minded and knowledgeable about alternative extensions will sell in .pro. If you registered a .pro in 2008 or 2009, it makes more sense to develop it.
In terms of other similar sales, Data.pro was caught on SnapNames for $725 in April 2008 and I picked up Database.pro, again on SnapNames, for $59. Datacenters.net sold for $1,300 in March 2008. I can't see you selling Datacenter.pro for more than $10-$20 on Namepros and even then you would struggle to sell it.
Beyond Sedo, you could try direct marketing to online datacenters but would you give up the accumulated backlinks and superior ageing of a bad 2 or 3 word .com and start against with a .pro when few people have heard of the extension?
The problem with .pro is because only a small number of people can register it, there are virtually no domainers to sell to. Volume .pro domainers either have their own collection of similar keywords that they can't sell or stronger keywords that they are fed up batting off lowball offers for.
Regging a .pro with a view to selling it on is like going for a walk in the desert without a map or a bottle of water, you aren't going to get far. You can either develop them into minisites and maybe cover your reg fee with some hard work or sit back and wait for RegistryPro's parent company Hostway to get fed up subsidising a loss making extension that has no chance of making any money until ICANN removes the restrictions on registering it.