公开(公告)号：US20130111475A1

公开(公告)日：2013-05-02

申请号：US13717463

申请日：2012-12-17

Applicant: VMware, Inc.

Inventor： Kinshuk GOVIL ,  Kiran KAMATH ,  Steven Michael KUSALO ,  Mark Russell JOHNSON

IPC: G06F9/455

CPC classification number: G06F9/455 ,  G06F9/5044

Abstract: A method of determining compatibility of a virtual machine or virtual machine disk file with a host, including a storage host, is disclosed. A lookup matrix is created to provide a fast compatibility lookup. To create the lookup matrix, computing object properties are retrieved from each of the plurality of computing objects. The computing object properties include resource requirements of each of the plurality of computing objects. Further, host properties are retrieved from each of the plurality of hosts. The method further includes creating host logical groups of a subset of the plurality of hosts having substantially same host properties and creating a plurality of computing object logical groups of a subject of the plurality of computing objects having substantially same computing object properties. The lookup matrix answers whether each member of a selected computing object logical group can be transferred to a selected host logical group.

Abstract translation: 公开了一种确定虚拟机或虚拟机磁盘文件与包括存储主机在内的主机的兼容性的方法。 创建查找矩阵以提供快速兼容性查找。 为了创建查找矩阵，从多个计算对象中的每一个检索计算对象属性。 计算对象属性包括多个计算对象中的每一个的资源需求。 此外，从多个主机中的每个主机检索主机属性。 该方法还包括创建具有基本相同的主机属性的多个主机的子集的主机逻辑组，并且创建具有基本相同的计算对象属性的多个计算对象的对象的多个计算对象逻辑组。 查找矩阵回答所选择的计算对象逻辑组的每个成员是否可以被传送到所选择的主机逻辑组。

公开(公告)号：US20130024862A1

公开(公告)日：2013-01-24

申请号：US13631447

申请日：2012-09-28

Applicant: VMWARE, INC.

Inventor： Beng-Hong LIM ,  Kinshuk GOVIL

IPC: G06F9/455

CPC classification number: G06F9/45558 ,  G06F8/656 ,  G06F9/5077 ,  G06F2009/4557

Abstract: In a virtualized system running one or more virtual machines on a first hypervisor, a second hypervisor is installed and control of the hardware resources of the physical computer supporting the virtualized system is migrated from the first hypervisor to the second hypervisor without interrupting the operation of the first hypervisor and the virtual machines. Initially a minimal set of hardware resources is hot-removed from control by the first hypervisor, and the second hypervisor is launched on the minimal set of hardware resources. Both the remaining hardware resources and the virtual machines are then migrated from the first hypervisor to the second hypervisor until all the virtual machines have been migrated over to the second hypervisor, while the virtual machines and the first hypervisor continue running largely unaffected by the migration process.

Abstract translation: 在运行第一管理程序上的一个或多个虚拟机的虚拟化系统中，安装第二管理程序，并且支持虚拟化系统的物理计算机的硬件资源的控制从第一管理程序迁移到第二管理程序，而不中断 第一个虚拟机管理程序和虚拟机。 最初，由第一虚拟机管理程序的控制热删除最少的一组硬件资源，并且在最小的硬件资源集上启动第二管理程序。 然后，剩余的硬件资源和虚拟机都从第一个虚拟机管理程序迁移到第二个虚拟机管理程序，直到所有的虚拟机已经迁移到第二个虚拟机管理程序，而虚拟机和第一个虚拟机监控程序继续运行在很大程度上不受迁移进程的影响 。

公开(公告)号：US20150058843A1

公开(公告)日：2015-02-26

申请号：US14311755

申请日：2014-06-23

Applicant: VMWARE, INC.

Inventor： Anne HOLLER ,  Jayanth GUMMARAJU ,  Kinshuk GOVIL ,  Benjamin J. CORRIE ,  George HICKEN

IPC: G06F9/455 ,  G06F9/50 ,  G06F9/48

CPC classification number: G06F9/455 ,  G06F9/45533 ,  G06F9/45558 ,  G06F9/48 ,  G06F9/50 ,  G06F9/5088 ,  G06F2009/4557

Abstract: A distributed computing application is described that provides a highly elastic and multi-tenant platform for Hadoop applications and other workloads running in a virtualized environment. Multiple instances of a distributed computing framework, such as Hadoop, may be executed concurrently. A centralized manager detects when contention for computing resources, such as memory and CPU, causes tasks to run slower on VMs executing on a given host, and scales up or scales down a cluster based on the detected resource contention.

Abstract translation: 描述了一种分布式计算应用程序，为Hadoop应用程序和在虚拟化环境中运行的其他工作负载提供了高弹性和多租户平台。 诸如Hadoop的分布式计算框架的多个实例可以同时执行。 集中式管理器检测在诸如内存和CPU之类的计算资源的竞争情况下，导致任务在给定主机上执行的VM上运行速度较慢，并且基于检测到的资源争用来扩展或缩小集群。