Fetching is an implementation of parallel predictive prefetching for Metropolis-Hastings. Prefetching algorithms use speculative execution to parallelize MCMC.
Elaine Angelino, Eddie Kohler, Amos Waterland, Margo Seltzer, Ryan P. Adams. Accelerating MCMC via parallel predictive prefetching. In 30th Conference on Uncertainty in Artificial Intelligence, UAI ’14, 2014.
Elaine Angelino. Accelerating Markov chain Monte Carlo via parallel predictive prefetching. PhD thesis, School of Engineering and Applied Sciences, Harvard University, 2014. Harvard version. Living version.
$ git clone git@github.com:elaine84/fetching.git
$ port install boost +openmpi
$ port install gsl
$ git clone git@github.com:yamins81/tabular.git
$ mpirun -np 2 ./fetching
$ mpirun -np 2 ./fetching -py pyblasso