That's the point where I cease to follow any discernible convention. I don't think I'm the only one.

  SQL:        "ess kyoo el"
  T-SQL:      "tee-sequel"
  PL/SQL:     "pee el ess kyoo el"
  SQLite:     "sequelite"
  SQL Server: "sequel server"
  MySQL:      "my ess kyoo el"
  PostgreSQL  "Postgres"

PostgreSQL may be the worst/best compromise. The Press FAQ provides an audio file the pronounces it post-grez-que-elle, keeping enough of both pronunciations to be a linguistic spork. Considering it's roots with Ingres, it makes sense, but it also feels like a layered pun.

http://www.postgresql.org/files/postgresql.mp3

It took me a while to learn "sequel" was the pronunciation, yes.

But this is because I learned SQL from reading about it, not going to conferences.

"SQL" is pretty complicated:

http://patorjk.com/blog/2012/01/26/pronouncing-sql-s-q-l-or-...

I say "olay" but "ess que ell". For me OLE is closer to an actual word (Olé), but SQL is nothing like "sequel".

SQL was originally SEQUEL, thus the pronunciation :)

Once interviewed a guy who kept talking about knowing "Squeel". Had no idea what he was talking about until he mentioned "Squeel Server". Cue banjo music...