CAJ | 학술논문

阐述了一种基于VMM(virtual machine manager)的虚拟机缓存划分的设计与实现.该方法采用操作系统中的页面着色技术,在虚拟机管理器Xen上进行实现.这种机制对于VMM之上的客户操作系统是完全透明的,便于操作,具有很好的灵活性.经测试表明,提出的缓存划分的方法能够显著地提高同时运行在不同虚拟机上的应用程序的性能.对从SPEc CPU 2006基准测试程序里面挑选出来的并发程序的负载进行测试,结果表明缓存划分最高可以使其性能提升19%.
천술료일충기우VMM(virtual machine manager)적허의궤완존화분적설계여실현.해방법채용조작계통중적혈면착색기술,재허의궤관리기Xen상진행실현.저충궤제대우VMM지상적객호조작계통시완전투명적,편우조작,구유흔호적령활성.경측시표명,제출적완존화분적방법능구현저지제고동시운행재불동허의궤상적응용정서적성능.대종SPEc CPU 2006기준측시정서리면도선출래적병발정서적부재진행측시,결과표명완존화분최고가이사기성능제승19%.
This paper presents a cache partitioning approach which can be implemented in the VMM (virtual machine manager). This approach has been implemented in Xen VMM using the page coloring technique traditionally applied to the OS (operating system). The VMM-based implementation is fully transparent to the guest OS. It thus shows the advantages of simplicity and flexibility. The evaluation shows that this cache partitioning method can work efficiently and improve the performance of co-scheduled applications running within different VMs. In the concurrent workloads selected from the SPEC CPU 2006 benchmarks, the technique achieves a performance improvement by up to 19% for the most sensitive workloads.