If the dimension of matrix X is a large number, can we use the RESCAL-ALS method?

[Jack-Ray]

Recently, I have been interested in the Tensor Decomposition. And I read your paper 'A Three-Way Model for Collective Learning on Multi-Relational Data'. Now I meet a problem that when dimension of matrix X is a large number, such as 10000X10000X11. It is too slow to use the RESCAL-ALS method to decompose the matrix. There was still no answer aftering running 6 hours, . So I want to know when the dimension of matrix is large, whether I have the efficient method to decompose the Matrix.