Random number generation in python for Kernighan–Lin and Simulated annealing algorithm?

Is it a good idea (In terms of quality of the number generated and CPU time required) to use Python's(2.7) default(Mersenne Twister) random() function as the random number generator for Kernighan–Lin algorithm? Is there some better way to do it?

Also, in same context, how would random() function work for generating numbers between 0 and 1 for Simulated Annealing algorithm?