This research aimed to investigate the efficacy of S-1 plus oxaliplatin (SOX) as perioperative chemotherapy for locally advanced gastric cancer. We enrolled 102 patients with preoperative clinical stage T3-4NxMO gastric cancer who were then randomly assigned to receive SOX as either perioperative chemotherapy (group A, 50 patients) or postoperative adjuvant chemotherapy (group B, 52 patients). Short-term curative efficacy and adverse effects of perioperative chemotherapy were analyzed. The rates of R0 resection, surgical complications, combined multiple organ resection, overall survival (OS), and disease-free survival (DFS) were compared between the groups. Results showed an overall response rate in group A of 42%, with a disease control rate of 94% and a tumor down-staging rate of 50%. An R0 resection rate of 90% was achieved in group A, which was significantly higher than that in group B (75%). No surgical mortality was observed, and the differences in surgical complications and combined multiple organ resection rates between the groups were not significant. The postoperative pathological examination of 4 patients in group A did not show any cancer cells in the tumor bed, resulting in a histological complete remission rate of 8%. The average OS and DFS for group A patients were 17.928 and 16.134 months, respectively, which were both longer than that of group B patients. However, the differences were not significant. In all, our results shows that in locally advanced gastric carcinoma, SOX perioperative chemotherapy is effective and results in a significantly improved R0 resection rate compared to postoperative SOX administration. Perioperative SOX does not cause additional surgical complications and has low adverse reaction rates; moreover, it appears to prolong survival.