With the widespread adoption of Internet of Things (IoT) technology across various sectors, leading to a substantial increase in terminal devices and task data volumes. Efficient task scheduling has thus become a critical challenge in cloud computing. To address this issue, this paper introduces a novel Dual Adaptive Genetic Algorithm (DAGA), building upon the original Adaptive Genetic Algorithm (AGA) but tailored for the evolving characteristics of cloud environments. DAGA not only prioritizes minimizing total task completion time but also emphasizes achieving balanced average task completion times. The study includes simulations in a cloud computing environment using Matlab, where DAGA is compared against AGA. Test parameters are carefully set and adjusted to evaluate and contrast the original and enhanced algorithms. Through rigorous testing, DAGA demonstrates superior performance over AGA in terms of both total job completion time and average task completion time.