How to parallelize the training of an SVC with RBF kernel through MapReduce in scikit-learn for Python

Question

How would I go about splitting up the training of a SVC classifier from the scikit-learn library among multiple processes (or completely separate computers)? I've found several papers that mention the feasibility of training an SVM iteratively in parallel, but I haven't found any specific examples that apply to the scikit-learn library.