Virtualization Support in Platform Management (PLM) Information Model

• US 20120233608A1
• Filed: 03/10/2011
• Published: 09/13/2012
• Est. Priority Date: 03/10/2011
• Status: Active Grant

• Current Assignee: Vivo Mobile Communication Co., Ltd. (Vivo Communication Technology Co. Ltd.)
• Original Assignee: Telefonaktiebolaget LM Ericsson
• Inventors: Toeroe, Maria
• Granted Patent: US 8,413,148 B2
• Time in Patent Office: Days
• US Class Current: 718/1
• CPC Class Codes: G06F 2009/4557 Distribution of virtual machine instances; Migration and load balancing
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects
• CPC Class Codes: G06F 9/5077 Logical partitioning of resources; Management or configuration of virtualized resources specific details on emulation or internal functioning of virtual machines G06F9/455

Abstract

A method and system for protecting against failure of hardware elements (HEs) in a virtual machine system using an information model with virtualization support. Based on the information model, the method boots virtual machine monitors (VMMs), virtual machines (VMs) and operating systems (OSes). The information model includes a graph of objects representing configured relationships between the hardware elements (HEs), VMMs, VMs, and OSes through parent child relationships between the graph'"'"'s objects. The graph of objects includes: a domain object, HE objects, VMM objects, VM objects, and OS objects. The VM objects of a migration-enabled type represent those of the VMs that may migrate between the VMMs. Based on the runtime associations between the VMM objects and the VM objects of the migration-enabled type, the method identifies which of the migration-enabled VMs are hosted by a same HE and are therefore vulnerable to hardware failure of the same HE.

Claims

1. A method implemented in a virtual machine system for protecting against failure of hardware elements (HEs), the HEs to host virtual machine monitors (VMMs), virtual machines (VMs) and operating system (OSes), each of the VMMs to manage one or more of the VMs, and each of the VMs to run one of the OSes, the method comprising the steps of:

booting, based on an information model of the virtual machine system, one of the VMMs on each of the HEs, one or more of the VMs, and one of the OSes on each of the VM, wherein the information model includes a graph of objects representing configured relationships between the HEs, VMMs, VMs, and OSes through parent child relationships between the graph'"'"'s objects, wherein a distinguished name (DN) is stored in each of the graph'"'"'s objects and those of the objects that are children of another object in the graph include the DN of their parent object thereby identifying the parent child relationships of the graph'"'"'s objects, wherein the graph of objects includes;

a domain object that represents the virtual machine system,

HE objects that represent the HEs and that are children of the domain object,

VMM objects that each represent one of the VMMs and that each are a child of one of the HE objects,

VM objects that each represent one of the VMs, wherein a set of one or more of the VM objects is of a migration-enabled type and represents those of the VMs that may migrate between the VMMs, and wherein the VM objects of the migration-enabled type are children of the domain object and not of the VMM objects such that their DNs do not change with migrations between the VMMs, and

OS objects that each represent one of the OSes and that each are a child of one of the VM objects or of the domain object,

wherein the step of booting includes the steps of

determining, for each of the VMs represented by the set of VM objects, a current management relationship that indicates which of the VMMs is to currently manage that VM; and

updating each of the set of VM objects to reflect the determined current management relationship as a runtime association; and

identifying, based on the runtime associations, which of the set of the VMs are hosted by a same HE of the HEs and are therefore vulnerable to hardware failure of the same HE.

2. The method of claim 1, wherein the step of booting further comprises the step of:

for a given one of the HEs, booting the highest boot ranking one of the VMMs represented by the VMM objects that are children of the HE object representing the given HE.

3. The method of claim 1, wherein the step of booting further comprises the step of:

for a given one of the HEs, booting the highest boot ranking one of the OSes represented by the OS objects that are children of the HE object representing the given HE.

4. The method of claim 1, wherein the step of booting further comprises the step of:

for a given one of the VMs, booting the highest boot ranking one of the OSes represented by the OS objects that are children of the VM object representing the given VM.

5. The method of claim 1, wherein the step of determining further comprises the step of:

determining, for each of the VMs represented by the set of VM objects, which of the VMMs is to currently manage that VM without exceeding a maximum number of VMs that can be managed by each of the VMMs.

6. The method of claim 1, further comprising the step of:

receiving a trigger to migrate a first of the VMs from a first of the HEs to a different one of the HEs, the first VM represented by a first of the set of VM objects that is of the migration-enabled type; and

selecting a preferred one of the VMMs to which to migrate the first VM based on a rank of the VMMs, the preferred VMM hosted by a second of the HEs.

7. The method of claim 6, wherein the step of updating further comprises the step of:

updating the runtime association of the first VM object to indicate that the first VM has migrated to the preferred VMM hosted by the second HE, wherein the first VM object is identified by the same DN, before and after any migration.

8. A computer system adapted to protect against failure of hardware elements (HEs) in a virtual machine system, the HEs to host virtual machine monitors (VMMs), virtual machines (VMs) and operating system (OSs), each of the VMMs to manage one or more of the VMs, and each of the VMs to run one of the OSes, the computer system comprising:

a processor coupled to a non-transitory computer readable storage medium, the non-transitory computer readable storage medium having stored therein a platform management (PLM) service module adapted to;

cause the virtual machine system to boot, based on an information model, one of the VMMs on each of the HEs, one or more of the VMs, and one of the OSes on each of the VM, wherein the information model includes a graph of objects representing configured relationships between the HEs, VMMs, VMs, and OSes through parent child relationships between the graph'"'"'s objects, wherein a distinguished name (DN) is stored in each of the graph'"'"'s objects and those of the objects that are children of another object in the graph include the DN of their parent object thereby identifying the parent child relationships of the graph'"'"'s objects, wherein the graph of objects includes;

a domain object that represents the virtual machine system,

HE objects that represent the HEs and that are children of the domain object,

VMM objects that each represent one of the VMMs and that each are a child of one of the HE objects,

VM objects that each represent one of the VMs, wherein a set of one or more of the VM objects is of a migration-enabled type and represents those of the VMs that may migrate between the VMMs, and wherein the VM objects of the migration-enabled type are children of the domain object and not of the VMM objects such that their DNs do not change with migrations between the VMMs, and

OS objects that each represent one of the OSes and that each are a child of one of the VM objects or of the domain object;

determine, for each of the VMs represented by the set of VM objects, a current management relationship that indicates which of the VMMs is to currently manage that VM; and

update each of the set of VM objects to reflect the determined current management relationship as a runtime association,

wherein the information model allows an availability management module to identify, based on the runtime associations, which of the set of the VMs are hosted by a same HE of the HEs and are therefore vulnerable to hardware failure of the same HE.

9. The computer system of claim 8, wherein the PLM service module is further adapted to:

for a given one of the HEs, boot the highest boot ranking one of the VMMs represented by the VMM objects that are children of the HE object representing the given HE.

10. The computer system of claim 8, wherein the PLM service module is further adapted to:

for a given one of the HEs, boot the highest boot ranking one of the OSes represented by the OS objects that are children of the HE object representing the given HE.

11. The computer system of claim 8, wherein the PLM service module is further adapted to:

for a given one of the VMs, boot the highest boot ranking one of the OSes represented by the OS objects that are children of the VM object representing the given VM.

12. The computer system of claim 8, wherein the PLM service module is further adapted to:

determine, for each of the VMs represented by the set of VM objects, which of the VMMs is to currently manage that VM without exceeding a maximum number of VMs that can be managed by each of the VMMs.

13. The computer system of claim 8, wherein the PLM service module is further adapted to:

receive a trigger to migrate a first of the VMs from a first of the HEs to a different one of the HEs, the first VM represented by a first of the set of VM objects that is of the migration-enabled type; and

select a preferred one of the VMMs to which to migrate the first VM based on a rank of the VMMs, the preferred VMM hosted by a second of the HEs.

14. The computer system of claim 13, wherein the PLM service module is further adapted to:

update the runtime association of the first VM object to indicate that the first VM has migrated to the preferred VMM hosted by the second HE, wherein the first VM object is identified by the same DN, before and after any migration.

15. A non-transitory computer readable storage medium including instructions that, when executed by a computer system, cause the computer system to perform a method of a virtual machine system for protecting against failure of hardware elements (HEs), the HEs to host virtual machine monitors (VMMs), virtual machines (VMs) and operating system (OSes), each of the VMMs to manage one or more of the VMs, and each of the VMs to run one of the OSes, the method comprising the steps of:

booting, based on an information model of the virtual machine system, one of the VMMs on each of the HEs, one or more of the VMs, and one of the OSes on each of the VM, wherein the information model includes a graph of objects representing configured relationships between the HEs, VMMs, VMs, and OSes through parent child relationships between the graph'"'"'s objects, wherein a distinguished name (DN) is stored in each of the graph'"'"'s objects and those of the objects that are children of another object in the graph include the DN of their parent object thereby identifying the parent child relationships of the graph'"'"'s objects, wherein the graph of objects includes;

a domain object that represents the virtual machine system,

HE objects that represent the HEs and that are children of the domain object,

VMM objects that each represent one of the VMMs and that each are a child of one of the HE objects,

VM objects that each represent one of the VMs, wherein a set of one or more of the VM objects is of a migration-enabled type and represents those of the VMs that may migrate between the VMMs, and wherein the VM objects of the migration-enabled type are children of the domain object and not of the VMM objects such that their DNs do not change with migrations between the VMMs, and

OS objects that each represent one of the OSes and that each are a child of one of the VM objects or of the domain object,

wherein the step of booting includes the steps of

determining, for each of the VMs represented by the set of VM objects, a current management relationship that indicates which of the VMMs is to currently manage that VM; and

updating each of the set of VM objects to reflect the determined current management relationship as a runtime association;

identifying, based on the runtime associations, which of the set of the VMs are hosted by a same HE of the HEs and are therefore vulnerable to hardware failure of the same HE.

16. The non-transitory computer readable storage medium of claim 15, where the step of booting further comprises the step of:

for a given one of the HEs, booting the highest boot ranking one of the VMMs represented by the VMM objects that are children of the HE object representing the given HE.

17. The non-transitory computer readable storage medium of claim 15, wherein the step of booting further comprises the step of:

for a given one of the HEs, booting the highest boot ranking one of the OSes represented by the OS objects that are children of the HE object representing the given HE.

18. The non-transitory computer readable storage medium of claim 15, wherein the step of booting further comprises the step of:

for a given one of the VMs, booting the highest boot ranking one of the OSes represented by the OS objects that are children of the VM object representing the given VM.

19. The non-transitory computer readable storage medium of claim 15, wherein the step of determining further comprises the step of:

determining, for each of the VMs represented by the set of VM objects, which of the VMMs is to currently manage that VM without exceeding a maximum number of VMs that can be managed by each of the VMMs.

20. The non-transitory computer readable storage medium of claim 15, wherein the method further comprises the step of:

receiving a trigger to migrate a first of the VMs from a first of the HEs to a different one of the HEs, the first VM represented by a first of the set of VM objects that is of the migration-enabled type; and

selecting a preferred one of the VMMs to which to migrate the first VM based on a rank of the VMMs, the preferred VMM hosted by a second of the HEs.

21. The non-transitory computer readable storage medium of claim 20, wherein the step of updating further comprises the step of:

updating the runtime association of the first VM object to indicate that the first VM has migrated to the preferred VMM hosted by the second HE, wherein the first VM object is identified by the same DN, before and after any migration.