The hardware solution consists of actually building specialized hardware capable of supporting WAM (or a similar low-level language). A considerable number of proposals have been made in the past for specialized hardware dedicated to support execution of logic programming languages.
The interest in hardware support for Prolog exploded mainly during the 1980's, and found a considerable source of inspiration in the work by Tick and Warren [109], where the design of a microcoded WAM is presented.
The experience of developing ad hoc hardware for logic programming languages, as happened for functional languages, has been only partially satisfactory. The experience of using sophisticate compiling techniques (as in Aquarius [90] and Parma [107]), and the high pace at which RISC technology is improving general-purpose processors speed, proved that there is nothing inherent in the Prolog language that prevents from executing it with speed at least comparable with that of imperative languages--without the need of explicit hardware support. This has emerged in the experience of Parma and Aquarius (by using processors with the same clock-rate, Parma running on a general-purpose processor is achieving a greater performance than VLSI-BAM--the specialized hardware designed for Aquarius).
In recent years the interest for special-purpose architectures has progressively died, and the work on hardware support for Prolog has been mainly redirected towards the design of the next generation of general purpose processors.