Vmem: A Lightweight Hot-Upgradable Memory Management for In-production Cloud Environment
Hao Zheng, Qiang Wang, Longxiang Wang, Xishi Qiu, Yibin Shen, Xiaoshe Dong, Naixuan Guan, Jia Wei, Fudong Qiu, Xingjun Zhang, Yun Xu, Mao Zhao, Yisheng Xie, Shenglong Zhao, Min He, Yu Li, Xiao Zheng, Ben Luo, Jiesheng Wu
传统的内存管理受到元数据开销、架构复杂性和稳定性下降的影响,这些问题在云环境中加剧。 现有的软件/硬件优化不足以满足云计算的灵活性和低开销的双重需求。 本文介绍了 Vmem,一种用于生产中的云环境的内存管理架构,通过轻量级保留内存管理实现灵活、高效的云服务器内存利用率。 Vmem是第一个支持在线升级的架构,满足云要求,以实现高稳定性和快速迭代演进。 实验表明,Vmem可将可销售内存速率提高约2%,提供极端弹性和性能,为基于VFIO的虚拟机(VM)实现超过3倍的启动时间,并将DPU加速的VM的网络性能提高了约10%。 Vmem已经大规模部署了七年,在支持数亿虚拟机的30多万台云服务器上展示了效率和稳定性。
Traditional memory management suffers from metadata overhead, architectural complexity, and stability degradation, problems intensified in cloud environments. Existing software/hardware optimizations are insufficient for cloud computing's dual demands of flexibility and low overhead. This paper presents Vmem, a memory management architecture for in-production cloud environments that enables flexible, efficient cloud server memory utilization through lightweight reserved memory management. Vmem i...
