An algorithm for online optimization of accelerators

A general algorithm is developed for online optimization of accelerator performance, i.e., online tuning, using the performance measure as the objective function. This method, named robust conjugate direction search (RCDS), combines the conjugate direction set approach of Powell's method with a robust line optimizer which considers the random noise in bracketing the minimum and uses parabolic fit of data points that uniformly sample the bracketed zone. It is much more robust against noise than traditional algorithms and is therefore suitable for online application. Simulation and experimental studies have been carried out to demonstrate the strength of the new algorithm.