To mitigate severe physical resource consumption in cloud data centers, we propose an Improved Sparrow Search Algorithm-Based Virtual Machine Placement (ISSA-VMP) method. Incorporating Chebyshev chaotic mapping and Levy flight disturbance enhances resource allocation diversity in the search space. The mapping encoding scheme transforms virtual machine placement solutions into continuous positional information. ISSA-VMP establishes a cloud data center resource consumption model to maximize physical host resource utilization efficiency. The simulation results demonstrate the excellent performance of ISSA-VMP in virtual machine migration and physical resource utilization, significantly reducing task completion time. Compared to the best performing algorithm, the execution rate has increased by 5.57–18.11%. ISSA-VMP achieves high and stable physical resource utilization rates, ensuring efficient utilization, with a stable Service Level Agreement (SLA) violation rate. In summary, ISSA-VMP is a promising, efficient solution for optimizing resource allocation in cloud data centers.

为了缓解云数据中心严重的物理资源消耗，我们提出了一种基于改进的 Sparrow 搜索算法的虚拟机放置（ISSA-VMP）方法。结合切比雪夫混沌映射和 Levy 飞行扰动增强了搜索空间中的资源分配多样性。映射编码方案将虚拟机放置解决方案转换为连续的位置信息。ISSA-VMP建立云数据中心资源消耗模型，最大化物理主机资源利用效率。仿真结果证明了ISSA-VMP在虚拟机迁移和物理资源利用率方面的优异性能，显着缩短了任务完成时间。与性能最好的算法相比，执行率提高了5.57-18.11%。ISSA-VMP实现了高且稳定的物理资源利用率，确保了高效利用，并具有稳定的服务级别协议（SLA）违反率。总之，ISSA-VMP 是一种有前景、高效的解决方案，用于优化云数据中心的资源分配。