This paper presents a joint optimization approach for microservice deployment and request routing in cloud data centers. The goal is to minimize the average response latency of the system under overhead constraints. The authors propose a hybrid genetic and local search-based deployment algorithm (HELAS) and a probabilistic forwarding-based routing algorithm (PORA). HELAS uses m/m/c queues to optimize service deployment and routing, while PORA considers the impact of instance location and computing resources on system performance to select optimal routing paths. The service graph with complex dependencies is decomposed into multiple service chains, and the open Jackson queuing network is applied to analyze the performance of the microservice system. Data evaluation results show that the proposed scheme significantly outperforms the benchmark strategy, reducing the average response latency by 37%-67% and enhancing the request success rate by 8%-115% compared to other baseline algorithms. The authors also develop a performance analysis model for service graphs with complex dependencies, utilizing the Open Jackson queuing network theory to conduct a comprehensive analysis of queuing delay, processing time limit, and data communication experiments. The joint optimization problem is NP-hard and resistant to traditional polynomial-time algorithms, so the authors propose a heuristic algorithm that leverages genetic and local search techniques to tackle the problem. The solution space is vast due to the abundance of microservices and their intricate interdependencies, requiring effective domain space definition, local optima avoidance, and acceleration of convergence. The authors' contributions include the development of a performance analysis model, the proposal of HELAS and PORA algorithms, and the application of queuing network analysis to optimize microservice deployment and request routing. The paper highlights the importance of considering both service deployment and request routing in the context of high-concurrency and massive internet application requests.This paper presents a joint optimization approach for microservice deployment and request routing in cloud data centers. The goal is to minimize the average response latency of the system under overhead constraints. The authors propose a hybrid genetic and local search-based deployment algorithm (HELAS) and a probabilistic forwarding-based routing algorithm (PORA). HELAS uses m/m/c queues to optimize service deployment and routing, while PORA considers the impact of instance location and computing resources on system performance to select optimal routing paths. The service graph with complex dependencies is decomposed into multiple service chains, and the open Jackson queuing network is applied to analyze the performance of the microservice system. Data evaluation results show that the proposed scheme significantly outperforms the benchmark strategy, reducing the average response latency by 37%-67% and enhancing the request success rate by 8%-115% compared to other baseline algorithms. The authors also develop a performance analysis model for service graphs with complex dependencies, utilizing the Open Jackson queuing network theory to conduct a comprehensive analysis of queuing delay, processing time limit, and data communication experiments. The joint optimization problem is NP-hard and resistant to traditional polynomial-time algorithms, so the authors propose a heuristic algorithm that leverages genetic and local search techniques to tackle the problem. The solution space is vast due to the abundance of microservices and their intricate interdependencies, requiring effective domain space definition, local optima avoidance, and acceleration of convergence. The authors' contributions include the development of a performance analysis model, the proposal of HELAS and PORA algorithms, and the application of queuing network analysis to optimize microservice deployment and request routing. The paper highlights the importance of considering both service deployment and request routing in the context of high-concurrency and massive internet application requests.