Most of the recent proposals in tabling technology were designed as a means to improve some practical deficiencies of current tabling execution models that reduce their applicability in particular applications. The discussion we address in this paper was also motivated by practical deficiencies we encountered, in particular, on the table storage mechanisms used for tabling support. To improve such mechanisms, we propose two new implementation techniques that make tabling models more efficient when dealing with incomplete tables and more robust when recovering memory from the table space. To validate our proposals, we have implemented them in the YapTab tabling system as an elegant extension of the original design.