Monte Carlo sampling (MCS) is the most widely used sampling technique. However, it is computationally expensive. Latin Hypercube sampling (LHS) is one of the well-known stratified sampling techniques. It has been found that LHS performs much better than MCS. However, LHS is not multi-dimensionally uniform, a desirable property for sampling. In 1997, we developed a sampling based on quasi-random sequences like Hammersley sequence sampling (HSS). HSS is k-dimensionally uniform and is very efficient sampling technique for solving problems of small dimensional uncertainties. The efficiency of this sampling technique for higher dimensions than 40 was improved by deriving a sampling called LHS-Hammersley sampling. This sampling technique performs well up to 250 dimensions. For higher dimensions than 250, we have another sampling technique called LHS-SOBOL. All these sampling techniques with large number of distribution functions are available as a tool from Stochastic Research Technologies LLC.