As Internet of things (IoT) has stepped out of the infancy and is gaining more and more popularity around the world, the gateway of environmental monitoring system has also always been attracting people’s eyes. With the capability to bridge application layer with sensing layer, gateway usually possesses multiple threads. Reasonably scheduling these threads matters most to the performance of gateway even the environmental monitoring system. Targeting at reasonable schedule, a kind of multiple threads schedule algorithm was researched and developed (MTSA) to weaken data jam by suitable duration of pending each thread and reasonable control for the execution sequence of thread. This duration was subject to the interval between two operations of pushing data from gateway to Web Server，the number of threads, and the relationship between maximal and effective power consumption. In order to overcome the difficulty in deciding the duration of pending each thread just by experience, the multiple threads integrated into the gateway of the environmental monitoring system were firstly displayed, and then the specific steps of MTSA were presented. In addition, the duration of pending each thread in MTSA was determined by taking the relationship between the maximal and effective power consumption into consideration, and the impact of MTSA on weakening data jam was also analyzed. With the purpose of demonstrating the performance of MTSA, a field experiment in the grape greenhouse was carried out. In this experiment, the comparison between MTSA and a popular algorithm of heuristic method (HM) was done in terms of decreasing the data loss to check the higher efficiency of MTSA than HM in guaranteeing that Web was capable of receiving the environment data, including temperature, relative humidity and illumine. As for the experiment results, data loss by NTSA was only 3.4%, which was much more less than that of 56% by HM, which verified the use of MSTA.