A typed dataset is similar to a dataset with only difference is that the sehema is already present in typed dataset. so if any mismatch in the column will generate compile time errors rather than runtime error as in the case of normal dataset. Also accessing the column value is much easier than the normal dataset as the column definition will be available in the schema.