Thanks to various observation campaigns over the past two decades, we have determined the most concordance Lambda-CDM cosmology parameters to sub-percent precision. The precision, however, does not guarantee the accuracy of the model. For example, the nature of 95 percent of the Lambda-CDM cosmology is unknown to us; hence, named dark matter and dark energy. The precision measurements also reveal tensions between cosmological parameters estimated from different probes, notably the Hubble parameter measured from the CMB versus the local observations. Again, the accurate cosmology model should come from observations, and the high-redshift (z>1) galaxy surveys are at the frontier. In this talk, I will show how we clarify the physical cosmology from large-scale-structure traced by high-redshift (z>1) galaxies and what is required to exploit the cosmological information buried in galaxies’ nonlinear clustering.