Map reduce to compute SVD (Singular value decomposition)

Is it possible to parallelize SVD computing, using for example Hadoop's MAP REDUCE?开发者_StackOverflow社区 Could you provide a simple example of it??

Yes, you can check Mahout Distributed Lanczos SVD implementation