Basically, any gene involved in the central processes of the cell (such as replication, transcription, RNA modification, translation, central metabolism etc) should be considered as house-keeping, which mounts their number to a few hundreds per bacterial genome, for instance. Although the precise limits of the "house-keeping genome" are not established, traditionally it comprises only those genes which directly perform or contribute to the central functions of life (invariant among most living organisms, making life as such, in a way) and does not include all sorts of "modulators" whose function is facultative, condition-specific or relates to a minor metabolic/regulatory pathway. House-keeping genes tend to be essential but are not always so. And, of course, not all of them are known (although it seems that we are aware of the absolute majority).