TRANSFORMING SET DATA-TYPES TO POWER OPTIMAL DATA-STRUCTURES

In this paper, we present a novel approach to model the search space f or the custom implementation of set data types, a data type that is co mmonly found in important application domains such as network componen t realizations and database applications. The main objective is to arr ive at power efficient realizations of these data types in custom data structures, but the model can also be used with nonpower cost functio ns. Based on the model, we propose an efficient optimization method fo r finding the implementation with minimum power consumption without pe rforming an exhaustive scan of the search space. The range of power co sts for different solutions can easily span four orders of magnitude, so a near optimal solution is crucial. This paper also strongly contri butes to our overall goal of a higher level of specification and short er design cycles for table-based memory organizations for applications where these data types are frequently used. The proposed model and me thodology are suited for both hardware and software realizations.