In the first part of the talk, I will focus on search algorithms in empirical performance tuning of computer codes. The increasing complexity, heterogeneity, and rapid evolution of modern computer architectures present obstacles for achieving high performance of scientific codes on different machines. Empirical performance tuning is a viable approach to obtain high-performing code variants based on their measured performance on the target machine. The search for the best code variant can be formulated as a numerical optimization problem. Two classes of algorithms are available to tackle this problem: global and local algorithms. I will present an experimental study of some global and local search algorithms on a number of problems from the recently introduced SPAPT test suite. The results show that local search algorithms are particularly attractive, where finding high-preforming code variants in a short computation time is crucial. In the second part of the talk, I will present the use of machine learning techniques to reduce computationally expensive simulations in chemical compound space and exascale workload characterization studies. Finally, I will give an overview of my future research plans.