That stuff about swap being twice the RAM is oudated information from the early days of BSD that somehow has survived as a sort of Urban legend in Linux. The reason for that rule was that early versions of BSD required that you have the same amount of swap as RAM for the system to run at all because when a program started, it geve it as much swap as RAM. Linux and most other modern operating systems treat swap like extra memory. So if you have 128M of swap plus 128M of RAM, it acts as though you have 256M of memory and swaps only when main memory runs low.
