Alvaro Herrera wrote:
Yeah, the problem seems underspecified.
So, Dan, the question is, what are you trying to measure? This might be a statistic that management has always been given, for Oracle, and you need to produce the "same" number for PostgreSQL. If not, it's hard to figure out what a statement counter actually can measure, to the extent that you can say, "If that number does THIS, I should do THAT." -- Engineers think that equations approximate reality. Physicists think that reality approximates the equations. Mathematicians never make the connection.
