System and method for adapting a system configuration of a first computer system for hosting on a second computer system

• US 8,533,305 B1
• Filed: 05/25/2012
• Issued: 09/10/2013
• Est. Priority Date: 09/23/2008
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: GoGrid LLC (Apollo Global Management, Inc.)
• Inventors: Carr, Jeffery, Keagy, John Martin, Lappas, Paul
• Primary Examiner(s): Boutah, Alina N
• Assistant Examiner(s): COONEY, ADAM A
• Application Number:
  US13/481,762
• Time in Patent Office: 473 Days
• Field of Search: 709/222, 718/1
• US Class Current: 709/222
• CPC Class Codes: G06F 2009/45562 Creating, deleting, cloning virtual machine instances
• CPC Class Codes: G06F 8/63 Image based installation; Cloning; Build to order
• CPC Class Codes: G06F 9/45533 Hypervisors; Virtual machine monitors
• CPC Class Codes: G06F 9/45537 Provision of facilities of other operating environments, e.g. WINE I/O emulation G06F13/105
• CPC Class Codes: G06F 9/4555 Para-virtualisation, i.e. guest operating system has to be modified
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects

Abstract

Some embodiments provide a method for a server hosting environment having several nodes. Each node is for using one or more hypervisors in order to host several configurations of several computer systems. The method receives a first configuration of a computer system. The first configuration includes an operating system that directly accesses hardware of the computer system. The method generates a second configuration that is based on the first configuration. The second configuration includes the first configuration modified to be operable on a particular hypervisor of a particular node in the several nodes. The method installs the second configuration on the particular node in order to allow the particular node to operate on the second configuration in conjunction with the particular hypervisor. The second configuration accesses a set of virtual hardware provided by the particular hypervisor.

Claims

1. A non-transitory computer readable storage medium storing a computer program for execution by at least one processor of a node of a server hosting environment comprising a plurality of nodes, each node for using one or more hypervisors in order to host a plurality of configurations of a plurality of computer systems, the computer program comprising:

a set of instructions for receiving a first configuration of a computer system, the first configuration comprising an operating system that uses a source kernel that interfaces directly with hardware of the computer system;

a set of instructions for automatically identifying a kernel that interfaces with virtual hardware provided by a particular hypervisor of the node;

a set of instructions for automatically modifying the operating system to be compatible with the identified kernel when the identified kernel is not compatible with the operating system; and

a set of instructions for installing on the node a second configuration comprising the operating system modified to use the identified kernel for interfacing with the virtual hardware provided by the particular hypervisor.

2. The non-transitory computer readable storage medium of claim 1, wherein the virtual hardware that interfaces with the identified kernel of the second configuration is a first set of virtual hardware, wherein a third configuration concurrently operates on the node and accesses a second, different set of virtual hardware provided by the particular hypervisor.

3. The non-transitory computer readable storage medium of claim 2, wherein the third configuration is based on a fourth configuration of a second, different computer system, wherein the third configuration comprises an operating system unmodified from the fourth configuration that is operable on the particular hypervisor.

4. The non-transitory computer readable storage medium of claim 2, wherein the third configuration comprises the same modified operating system as the first configuration, wherein the modified operating system of the third configuration uses the same identified kernel as the second configuration, said kernel of the third configuration for interfacing with the second set of different virtual hardware provided by the particular hypervisor.

5. The non-transitory computer readable storage medium of claim 2, wherein the particular hypervisor comprises (i) a first mapping of the first set of virtual hardware to a first set of physical hardware components of the node and (ii) a second mapping of the second set of virtual hardware to a second set of physical hardware components of the node, wherein a first virtual hardware component of the first set of virtual hardware is mapped to the same particular physical hardware component to which a second, different virtual hardware component of the second set of virtual hardware is mapped.

6. For a server hosting environment comprising a plurality of nodes, wherein a particular node comprises a hypervisor for hosting one or more computer system configurations, a method comprising:

receiving a configuration comprising a kernel, an operating system, and a set of device drivers, the configuration for directly accessing hardware of a computer system;

when the kernel is not compatible with virtual hardware provided by the hypervisor of the particular node, retrieving information for a compatible destination kernel from a kernel library storing different kernels for interfacing with different hardware on different platforms;

when one or more of the device drivers is not compatible with the destination kernel or a virtual hardware device of the hypervisor corresponding to the device driver, retrieving, for each incompatible device driver, information for a compatible destination device driver from a driver library storing different device drivers for interfacing with different kernels and different hardware on different platforms;

when the operating system is not compatible with the destination kernel, retrieving information for a compatible operating system from an operating system library storing different operating systems for interfacing with different kernels that access different hardware on different platforms; and

adapting the configuration with the retrieved information in order for the adapted configuration to be operable on the hypervisor of the particular node for hosting the configuration within the server hosting environment.

7. The method of claim 6, wherein the adapted configuration uses the destination kernel which is retrieved from the kernel library and is different from the kernel of the received configuration in order to access the virtual hardware provided by the hypervisor of the particular node.

8. The method of claim 7, wherein the operating system of the received configuration is compatible with the destination kernel retrieved from the kernel library.

9. The method of claim 7, wherein the operating system of the received configuration is not compatible with the destination kernel retrieved from the kernel library, wherein the adapted configuration uses the compatible operating system which is retrieved from the operating system library.

10. The method of claim 6, wherein each device driver of the received configuration directly interfaces with a physical hardware component of the computer system, wherein the adapted configuration uses a set of destination device drivers retrieved from the driver library, wherein each destination device driver of the adapted configuration directly interfaces with a virtual hardware component that is mapped to a physical hardware component of the particular node.

11. For a server hosting environment comprising a plurality of nodes, each node for using one or more hypervisors in order to host a plurality of configurations of a plurality of computer systems, a method comprising:

receiving a first configuration of a computer system, the first configuration comprising an operating system that directly accesses hardware of the computer system with a set of device drivers compatible with the operating system, each device driver corresponding to a hardware device of the computer system;

for a particular destination node in the plurality of nodes, automatically identifying a virtual hardware device provided by a particular hypervisor that does not correspond to any device driver in the set of device drivers of the first configuration;

automatically replacing a device driver in the set of device drivers with a compatible device driver that corresponds to the virtual hardware device, the replaced device driver incompatible with the virtual hardware device; and

installing on the particular destination node a second configuration comprising the operating system and the compatible device driver in order to allow the second configuration to operate on the particular destination node using the compatible device driver to access the virtual hardware device provided by the particular hypervisor.

12. The method of claim 11, wherein the first configuration further comprises a first kernel that interfaces with the operating system of the first configuration and the hardware of the computer system, wherein the method further comprises modifying the first kernel in order to generate a second kernel that interfaces with the operating system and a set of virtual hardware provided by the particular hypervisor.

13. The method of claim 11, wherein the particular hypervisor provides a set of virtual hardware devices that each corresponds to a device driver in the set of device drivers of the first configuration, wherein the second configuration further comprises the set of device drivers to allow the operating system of the second configuration to access the set of virtual hardware devices.

14. The method of claim 13, wherein the particular hypervisor comprises a mapping of the set of virtual hardware devices to a set of physical hardware devices of the particular destination node.

15. The method of claim 11, wherein automatically replacing the device driver comprises:

determining, among a particular set of device drivers stored in a driver library, whether any device driver in the driver library is compatible with the operating system and the virtual hardware device; and

retrieving, when a particular device driver in the driver library is determined to be compatible, the particular compatible device driver from the driver library.

16. The method of claim 15, wherein automatically replacing the device driver further comprises generating, when no device driver in the driver library is determined to be compatible, the particular compatible device driver using one of a kernel driver API and a development environment.

17. A method comprising:

remotely accessing a computer system;

from the computer system, extracting a configuration comprising an operating system and a set of device drivers that directly access a first set of physical hardware of the computer system;

generating a first configuration package comprising the operating system and the set of device drivers of the configuration; and

providing the first configuration package to an adaptation engine that (1) creates a second configuration package that is a modified version of the first configuration package with different device drivers for accessing a set of virtual hardware corresponding to a second set of physical hardware of a node of a hosting environment that comprises a plurality of nodes for hosting a plurality of configurations, and (2) supplies the second configuration package to the node in order to allow the node to host the configuration.

18. The method of claim 17, wherein the computer system is accessed remotely from a node of the hosting environment by logging into the computer system using a set of predefined log-in parameters.

19. The method of claim 17, wherein extracting the configuration comprises analyzing a file system of the computer system to identify the operating system and the set of device drivers.

20. The method of claim 17, wherein the plurality of nodes form a grid of nodes over which the plurality of configurations are hosted, wherein a particular configuration in the plurality of configurations is logically hosted by a set of two nodes in the grid of nodes.