Method and apparatus for providing virtual computing services

• US 20050044301A1
• Filed: 04/26/2004
• Published: 02/24/2005
• Est. Priority Date: 08/20/2003
• Status: Abandoned Application

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Davis, Scott Howard, Vasilevsky, Alexander David, Thomas, Benjamin Joseph III
• Application Number:
  US10/831,973
• Publication Number:
  US 20050044301A1
• Time in Patent Office: Days
• US Class Current: 711/1
• CPC Class Codes: G06F 9/45533 Hypervisors; Virtual machine monitors
• CPC Class Codes: G06F 9/50 Allocation of resources, e.g. of the central processing unit [CPU]
• 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
• CPC Class Codes: G06F 9/5083 Techniques for rebalancing the load in a distributed system

Abstract

A level of abstraction is created between a set of physical processors and a set of virtual multiprocessors to form a virtualized data center. This virtualized data center comprises a set of virtual, isolated systems separated by a boundary referred as a partition. Each of these systems appears as a unique, independent virtual multiprocessor computer capable of running a traditional operating system and its applications. In one embodiment, the system implements this multi-layered abstraction via a group of microkernels, each of which communicates with one or more peer microkernel over a high-speed, low-latency interconnect and forms a distributed virtual machine monitor. Functionally, a virtual data center is provided, including the ability to take a collection of servers and execute a collection of business applications over a compute fabric comprising commodity processors coupled by an interconnect. Processor, memory and I/O are virtualized across this fabric, providing a single system, scalability and manageability. According to one embodiment, this virtualization is transparent to the application, and therefore, applications may be scaled to increasing resource demands without modifying the application.

Claims

1. A system comprising:

a processor;

an operating system that accesses one or more virtual resources; and

an abstraction layer that is adapted to map the one or more virtual resources to one or more physical resources.

2. The system according to claim 1, wherein the one or more physical resources includes at least one of input/output (I/O) devices, processors, interconnects and storage entities.

3. The system according to claim 1, wherein the operating system is adapted to generate a plurality of function calls, and wherein the abstraction layer is adapted to map the plurality of function calls into processor instructions to be executed on a processor.

4. The system according to claim 1, wherein at least one of the operating system and a user program is adapted to generate an exception, and wherein the abstraction layer is adapted to handle the generated exception.

5. The system according to claim 1, wherein the abstraction layer is adapted to handle one or more traps generated by at least one of an operating system and a user program.

6. The system according to claim 1, wherein the operating system is adapted to generate a plurality of function calls, and wherein the abstraction layer is adapted to rewrite the function calls as processor instructions to be executed on a processor.

7. The system according to claim 6, wherein the abstraction layer is adapted to rewrite the plurality of function calls during execution of the plurality of function calls.

8. The system according to claim 1, wherein the abstraction layer presents a virtual hardware interface to the operating system.

9. The system according to claim 1, wherein the abstraction layer includes a microkernel that executes on a node.

10. The system according to claim 9, wherein the node includes one or more physical processors.

11. The system according to claim 8, wherein the abstraction layer presents an interface of a virtual processor to the operating system.

12. The system according to claim 8, wherein a plurality of virtual processors is grouped into a virtual system to present the virtual hardware interface.

13. The system according to claim 9, wherein the abstraction layer includes a plurality of microkernels, at least one of which operates cooperatively to share resources of the system and form a distributed virtual machine monitor.

14. The system according to claim 13, wherein the at least two of the plurality of microkernels cooperate to present a logical address space to the operating system through a virtual hardware interface.

15. The system according to claim 9, wherein the distributed virtual machine monitor is adapted to schedule a plurality of tasks as a plurality of respective threads executed by one or more of the plurality of physical processors.

16. The system according to claim 9, wherein the distributed virtual machine monitor is adapted to schedule a plurality of processes as a plurality of respective threads executed by one or more of the plurality of physical processors.

17. The system according to claim 15, wherein the plurality of threads share one or more objects via the operating system.

18. The system according to claim 17, wherein the operating system executes on a virtual processor.

19. The system according to claim 1, wherein the system further comprises a plurality of nodes, and wherein the processor is located in at least one of the plurality of nodes.

20. The system according to claim 1, wherein the abstraction layer defines a plurality of virtual processors, at least two of which are mapped by the abstraction layer to at least two physical processors, respectively, residing within a single node.

21. The system according to claim 19, wherein the abstraction layer comprises a plurality of microkernels, at least two of which operate cooperatively as a cluster.

22. The system according to claim 1, wherein the system further comprises a node, and wherein the processor is located in the node.

23. The system according to claim 22, wherein a microkernel program is executed by the node.

24. The system according to claim 21, wherein the plurality of microkernels forming a distributed virtual machine monitor that presents a single hardware interface to the operating system.

25. The system according to claim 1, wherein the abstraction layer emulates a processor architecture.

26. The system according to claim 25, wherein the processor architecture is an architecture of a virtual processor.

27. The system according to claim 26, wherein at least one of a plurality of function calls in the virtual processor architecture are mapped to one or more physical processors.

28. The system according to claim 11, wherein the operating system is adapted to execute on the virtual processor.

29. The system according to claim 1, wherein the abstraction layer presents a single virtual system to the operating system, and wherein the abstraction layer maps the virtual system onto one or more of a plurality of physical processors.

30. The system according to claim 12, wherein the number of virtual processors is permitted to be at least one of:

less than or equal to a number of physical processors in the system; and

any number in relation to the number of physical processors in the system.

31. The system according to claim 1, wherein the abstraction layer presents to the operating system one or more virtual processors, wherein the one or more virtual processors are mapped onto a plurality of physical processors.

32. The system according to claim 31, wherein one or more of the plurality of physical processors are hardware processors that reside within a node.

33. The system according to claim 31, wherein one or more virtual processors are associated with a virtual server, and wherein the one or more virtual processors are mapped onto one or more physical processors that reside within at least one of a group comprising a single node and different nodes.

34. The system according to claim 31, wherein at least one of the one or more virtual processors is associated with a first virtual server, and another one of the one or more virtual processors is associated with a second virtual server, and wherein the at least one of the virtual processors and the another one of the one or more virtual processors are mapped to respective physical processors that reside within a single node.

35. The system according to claim 9, wherein the abstraction layer is adapted to define a virtual server system comprising one or more virtual processors, and wherein the system further comprises a hierarchical scheduling system that allows a plurality of virtual processors to be shared among the plurality of physical processors.

36. The system according to claim 35, wherein the hierarchical scheduling system comprises a first and second scheduler, wherein the operating system includes the first scheduler which schedules at least one task to be performed by the virtual server system, and wherein the distributed virtual machine monitor includes the second scheduler, the second scheduler being associated with the virtual server system and being adapted to schedule the at least one task to be executed by at least one physical processor associated with the virtual server.

37. The system according to claim 12, wherein the system comprises virtual memory shared among the plurality of virtual processors.

38. The system according to claim 37, wherein each of the plurality of virtual processors is restricted to use a distributed memory that is mapped onto one or more physical memory locations.

39. The system according to claim 1, wherein the abstraction layer presents to the operating system a virtual cache-coherent, non-uniform memory access (NUMA) system.

40. The system according to claim 39, wherein the virtual NUMA system comprises a set of virtual processors executing on one or more physical nodes.

41. The system according to claim 39, wherein the virtual NUMA system provides access to basic I/O, memory and processor abstractions.

42. The system according to claim 1, wherein the abstraction layer presents to the operating system a virtual cache-only memory architecture (COMA).

43. The system according to claim 42, wherein the virtual COMA system comprises a set of virtual processors executing on one or more physical nodes.

44. The system according to claim 42, wherein the virtual COMA system provides access to basic I/O, memory and processor abstractions.

45. The system according to claim 1, wherein the abstraction layer presents to the operating system a virtual uniform memory access (UMA) architecture.

46. The system according to claim 37, wherein the virtual UMA system comprises a set of virtual processors executing on one or more physical nodes.

47. The system according to claim 37, wherein the virtual NUMA system provides access to basic I/O, memory and processor abstractions.

48. The system according to claim 9, wherein the distributed virtual machine monitor isolates the operating system from one or more physical processors.

49. The system according to claim 9, wherein the distributed virtual machine monitor isolates the operating system from a shared virtual memory system.

50. The system according to claim 13, wherein at least two of the plurality of microkernels are adapted to communicate using a synchronization protocol.

51. The system according to claim 50, wherein at least two of the plurality of microkernels communicate using a cache-only distributed shared memory paging protocol.

52. The system according to claim 50, wherein at least two of the plurality of microkernels communicate using a function shipping protocol.

53. The system according to claim 51, wherein the function shipping protocol is an object-level function shipping protocol.

54. The system according to claim 50, wherein at least two of the plurality of microkernels communicate using at least one of distributed shared memory protocol and a function shipping protocol.

55. The system according to claim 53, wherein the function shipping protocol is an object-level function shipping protocol.

56. The system according to claim 50, wherein at least two of the plurality of microkernels share objects.

57. The system according to claim 50, wherein at least two of the plurality of microkernels share pages.

58. The system according to claim 50, wherein at least two of the plurality of microkernels share objects and pages.

59. The system according to claim 8, wherein the abstraction layer presents, to one or more operating systems a plurality of virtual processors grouped into one or more virtual multiprocessor systems.

60. The system according to claim 59, wherein each of the one or more operating systems are executed by a respective one of the one or more virtual multiprocessor systems.

61. The system according to claim 59, wherein the system is adapted to migrate physical processors among the one or more virtual multiprocessor systems.

62. The system according to claim 59, wherein the one or more of virtual multiprocessor systems span one or more physical processors.

63. The system according to claim 59, wherein at least one virtual multiprocessor system may be configured to include or exclude one or more virtual processors while the virtual multiprocessor system is in an operating mode.

64. The system according to claim 14, wherein the at least one memory address space is a physical address space.

65. A computer-readable medium having stored thereon a data structure comprising data mapping a virtual processor to one or more physical processors.

66. A system comprising:

a manager adapted to create an instance of a virtual server; and

a resource allocated to the instance of the virtual server, the virtual resource being associated with one or more actual resources.

67. The system according to claim 66, wherein the resource includes at least one virtual processor being associated with one or more nodes.

68. The system according to claim 66, wherein the virtual server is associated with multiple virtual processors.

69. The system according to claim 67, wherein the virtual server presents a single system image while executing across the one or more nodes.

70. The system according to claim 67, wherein the one or more nodes are coupled by an interconnect.

71. The system according to claim 70, wherein the interconnect is InfiniBand.

72. The system according to claim 70, wherein the interconnect is PCI-Express.

73. The system according to claim 70, wherein the interconnect is GigaNet.

74. The system according to claim 70, wherein the interconnect is Gigabit Ethernet.

75. The system according to claim 70, wherein the interconnect is 10 Gigabit Ethernet.

76. The system according to claim 70, wherein the interconnect uses RDMA.

77. The system according to claim 67, wherein the virtual server is adapted to execute an application on the one or more nodes without modification.

78. The system according to claim 66, wherein the manager is adapted to perform a mapping between the virtual resource and the one or more actual resources.

79. The system according to claim 78, wherein the resource is a virtual network interface, and the actual resource includes an actual network interface.

80. The system according to claim 78, wherein the virtual resource is a virtual storage entity, and the actual resource includes an actual storage device.

81. The system according to claim 78, wherein the manager is adapted to perform a reallocation of the virtual resource to another virtual server instance.

82. The system according to claim 67, wherein the system further comprises a distributed server that is executed on the one or more nodes.

83. The system according to claim 82, wherein the manager is adapted to perform an association between the one or more actual resources and at least one of the distributed server and virtual server.

84. The system according to claim 82, wherein the system includes one or more unallocated actual resources, and wherein the manager is adapted to perform an association between the one or more unallocated resources and one or more respective distributed servers or virtual servers.

85. The system according to claim 66, wherein the virtual server is associated with a first collection of resources, and wherein the manager is adapted to disassociate the virtual server with the first collection of resources, and wherein the manager is adapted to associate the virtual server with a second collection of resources.

86. The system according to claim 85, wherein the second collection of resources is presented to the virtual server by a distributed server.

87. The system according to claim 86, wherein the distributed server is executed on by plurality of nodes.

88. A computer system comprising:

a plurality of processors; and

a virtualization layer adapted to define one or more virtual servers, at least one of which presents a single computer system interface to an operating system, the single computer system interface defining a plurality of instructions, wherein at least one of the plurality of instructions is directly executed on at least one of the plurality of processors, and at least one other of the plurality of instructions is handled by the virtualization layer.

89. The computer system according to claim 88, wherein the virtualization layer includes a microkernel that executes on at least one processor.

90. The computer system according to claim 88, wherein the at least one of the plurality of instructions is a non-privileged instruction.

91. The computer system according to claim 88, wherein the at least one other of the plurality of instructions is a privileged instruction.

92. The computer system according to claim 91, wherein the virtualization layer includes code that handles a call to the privileged instruction.

93. The computer system according to claim 88, wherein the virtualization layer passes the non-privileged instruction to the at least one of the plurality of processors without intervention.

94. The computer system according to claim 88, further comprising a plurality of resources, wherein each of the plurality of processors executes a respective instance of a microkernel program, and wherein each of the respective instances of the microkernel program are adapted communicate to cooperatively share the plurality of resources of the computer system.

95. The computer system according to claim 88, wherein the virtual server includes one or more virtual processors, wherein the virtualization layer is adapted to schedule tasks associated with at least one of the one or more virtual processors as a thread that is executed on at least one of the plurality of processors.

96. The computer system according to claim 95, wherein the virtualization layer is adapted to schedule a plurality of virtual processor tasks for execution substantially in parallel.

97. The computer system according to claim 88, further comprising a plurality of resources, wherein at least one of the one or more virtual servers includes at least two virtual interfaces, both of which are adapted to send requests for access to the plurality of resources in parallel.

98. The computer system according to claim 88, wherein at least one of the at least two virtual interfaces includes a virtual network interface.

99. The computer system according to claim 88, wherein at least one of the at least two virtual interfaces includes a virtual storage adapter.

100. The computer system according to claim 99, wherein the virtual storage adapter is a virtual host bus adapter (HBA).

101. The computer system according to claim 97, further comprising at least one I/O server, wherein the parallel access requests are serviced in parallel by the I/O server.

102. The computer system according to claim 97, further comprising at least one I/O device, wherein the parallel access requests are serviced in parallel by the I/O device.

103. The computer system according to claim 97, wherein the parallel access requests are transmitted over a switched communication network.

104. The computer system according to claim 103, wherein the switched communication network includes an InfiniBand switched fabric.

105. The computer system according to claim 97, wherein the parallel access requests are transmitted over a packet-based network.

106. The computer system according to claim 88, wherein the virtualization layer is adapted to map one or more virtual resources to one or more physical resources.

107. The computer system according to claim 106, wherein the one or more physical resources includes at least one of input/output (I/O) devices, processors, interconnects and storage entities.