The response time of Server 1 is $2x$ ms under load $x$ (queries/sec), and Server 2 is $4y$ ms under load $y$. If the total load is $x+y=30$, what load $x$ on Server 1 minimizes the total query latency (defined as $x T_1(x) + y T_2(y)$)?