Finding optimal quantum annealer architectures

Quantum annealing is a promising method for finding the global minimum of a given cost function with many instances, particularly useful for search spaces with a large number of local minima, like the energy of a spin glass.\\
But what architecture should a quantum annealer be equipped with?\\
Since we expect quantum annealing to give us an edge over classical simulated annealing, the search is for those architectures which are hard to solve classically. Because solving these currently requires extensive resources, being able to arrive at their solution quickly would provide the greatest gains.\\
By building and exploring architecutres that are increasingly harder to solve classically, we aim fo find their common characteristics and discover what makes a good quantum annealer architecture.