Notice: For security purposes, we have sent the file(s) attached to this e-mail in encrypted ZIP format. You will receive a separate e-mail containing a password to the file(s). Hello! Here is the data you asked for, plus I added the total record count for the table. Some of the column names are a bit different from what I posted before because I changed them for readability ss <-> sessions med <-> medium src <-> source (etc) I could not get a result for one question, "select count(distinct src)" after eight hours, so this may well be a big part of the problem. I got a value for that question by parsing the raw data. FYI this is a table of Internet traffic that's been analyzed in various ways. One possibility for the root problem may be that there was a bug (https://www.monetdb.org/bugzilla/show_bug.cgi?id=3646) in Oct2014-SP1 concerning the ordering of the 'inet' datatype. Since the search to see if a given 'src' is DISTINCT probably relies on sorting at some level, I wondered if upgrading would solve the problem. Unfortunately it does not seem so. A query for count(distinct src) done after the upgrade has been running for an hour with no results and a lot of disk write operations. However running a test case I find that there is still a problem (just filed as https://www.monetdb.org/bugzilla/show_bug.cgi?id=3660) with comparison operators on the 'inet' datatype, so there may yet be a core issue with handling that datatype that makes "count(distinct...)" inefficient. If there is any other data I can provide, or tests I could run, just let me know, and thanks for your help! Tim