Many networks are characterized by highly heterogeneous distributions of links which are called scale-free networks, and the degree distributions follow p ( k ) ∼ ck - α . We study the robustness of scale-free networks to random failures from the character of their heterogeneity. Entropy of the degree distribution can be an average measure of a network's heterogeneity. Optimization of scale-free networks’ robustness to random failures with average connectivity constant is equivalent to maximizing the entropy of the degree distribution. By examining the relationship of the entropy of the degree distribution, scaling exponent and the minimal connectivity, we get the optimal design of scale-free networks to random failures. We conclude that the entropy of the degree distribution is an effective measure of network's resilience to random failures.