In this paper, we investigate resource allocation in a mobile edge computing (MEC) network comprising multiple collaborative MEC servers. Our scenario assumes that when a server’s computational abilities are insufficient for a user’s task, it can delegate a portion of the workload to other servers. Effectively addressing the cooperative resource allocation challenge associated with such task offloading is the primary objective of our work. Specifically, we aim to minimize the overall energy consumption of all users while satisfying task latency constraints, user transmission rate limits, and servers’ computational capacity bounds. To overcome this challenge, we propose an iterative algorithm leveraging a combination of alternating optimization and convex optimization techniques. In alternating optimization, we divide the variables into two groups and optimize them alternatively while fixing the other group. For the inner convex optimization problems, we employ dual decomposition to decompose the original problem into independent sub- problems. Extensive simulation experiments were conducted to evaluate our proposed algorithm’s performance under varying network conditions. The results demonstrate that it outperforms both non-cooperative approaches as well as other benchmark cooperative algorithms, achieving superior energy efficiency and lower overall latency. Looking ahead, we plan to investigate more sophisticated workload partitioning and server cooperation schemes, as well as extending our approach to multipath transmission scenarios.