In this paper, we report our recent advancements in the area of iterative learning based control and optimization for large scale systems. Iterative learning control (ILC) for large scale systems consists of two categories. One is that the subsystems are physically interconnected and each subsystem has its own control objective. The other is that each subsystem is isolated, but the control objective is defined at the group level. Two sets of control schemes are described to solve these types of problems respectively. Whereas, in the area of iterative learning (IL) based optimization, parameter optimization and random perturbation based searching algorithms are presented. Finally, application examples in multi-agent systems control, power network dispatch, and freeway traffic network scheduling are discussed to demonstrate the merits of iterative learning methods.