IP Library Granted Patent US 8,370,530
Granted Patent B2
US 8,370,530 · App. 11/953,832 · Granted Feb 5, 2013

Method and system for controlling network traffic in a blade chassis

Inventors: Sunay Tripathi (San Jose, CA); Erik Nordmark (Mountain View, CA)
Assignee: Oracle America, Inc.
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 8,370,530
App. No.
11/953,832
Granted
Feb 5, 2013
Kind
B2
Abstract

In general, embodiments of the invention relates to a method for controlling network traffic in a chassis. The method includes assigning control of a network express manager located in the chassis to a control virtual machine selected from a number of virtual machines. The method further includes configuring the network express manager, by the control virtual machine, where the network express manager is configured to route network traffic in the chassis. The method further includes implementing a virtual network path using the network express manager, where the virtual network path includes a virtual wire between a first VNIC and a second VNIC, where the first VNIC is located in a first computer and the second VNIC is located in a second computer.

Claims (69)

1. A method for controlling network traffic in a chassis, comprising:

assigning control of a network express manager located in the chassis to a control virtual machine selected from a plurality of virtual machines,

wherein the plurality of virtual machines is located in a plurality of computers communicatively coupled with each other via a chassis interconnect in the chassis, wherein the chassis interconnect comprises a Peripheral Component Interface Express (PCI-E) backplane comprising a plurality of PCI-E endpoints, wherein a first computer of the plurality of computers connects to the chassis using a first PCI-E endpoint from the plurality of PCI-E endpoints, wherein a second computer of the plurality of computers connects to the chassis using a second PCI-E endpoint from the plurality of PCI-E endpoints,

wherein each the plurality of virtual machines is associated with at least one of a plurality of virtual network interface cards (VNICs),

wherein the plurality of computers share a physical network interface, and

wherein assigning control of the network express manager to the control virtual machine comprises mapping a service processor located in the network express manager as a physical device in the control virtual machine and wherein mapping the service processor is performed by a chassis processor located in the chassis;

configuring the network express manager, by the control virtual machine, wherein the network express manager is configured to route network traffic in the chassis;

implementing a virtual network path using the network express manager, and

migrating, after implementing the virtual network path, the control virtual machine within the plurality of computers without disrupting operation of the network express manager,

wherein the virtual network path comprises a virtual wire between a first VNIC selected from the plurality of VNICs and a second VNIC selected from the plurality of VNICs,

wherein the network express manager implements the virtual network path using a virtual switching table,

wherein the virtual switching table comprises a first mapping of a first Internet Protocol (IP) address for the first VNIC to the first PCI-E endpoint, and a second mapping of a second IP address for the second VNIC to the second PCI-E endpoint,

wherein the first VNIC is located in the first computer, and

wherein the second VNIC is located in the second computer.

2. The method of claim 1 , further comprising:

detecting a fault in the first computer, wherein migrating the control virtual machine is initiated based on the fault.

3. The method of claim 1 , further comprising:

checkpointing a control thread in the control virtual machine prior to migrating the control virtual machine; and

resuming the control thread after migrating the control virtual machine.

4. The method of claim 1 , wherein after mapping the service processor the service processor is not visible to any of the plurality of virtual machines other than the control virtual machine.

5. The method of claim 1 , wherein each of the plurality of computers is a blade.

6. A blade chassis comprising:

a chassis interconnect comprising a Peripheral Component Interface Express (PCI-E) backplane comprising a plurality of PCI-E endpoints;

a physical network interface;

a network express manager comprising a service processor;

a plurality of blades communicatively coupled with each other via the chassis interconnect,

wherein the plurality of blades shares the physical network interface,

wherein the plurality of blades comprises a plurality of virtual machines, and

wherein the plurality of virtual machines is associated with a plurality of virtual network interface cards (VNICs); and

a chassis processor configured to:

assign control of the network express manager to a control virtual machine selected from the plurality of virtual machines, wherein assigning control of the network express manager to the control virtual machine comprises mapping the service processor located in the network express manager as a physical device in the control virtual machine and wherein mapping the service processor is performed by a chassis processor located in the chassis,

migrate within the plurality of blades without disrupting operation of the network express manager,

wherein controlling the network express manager, by the control virtual machine, comprises configuring the network express manager to route network traffic in the blade chassis, and

wherein network traffic is routed in the blade chassis over a virtual network path,

wherein the network express manager implements the virtual network path using a virtual switching table,

wherein the virtual switching table comprises a first mapping of a first Internet Protocol (IP) address for the first VNIC to a first PCI-E endpoint, and a second mapping of a second IP address for the second VNIC to a second PCI-E endpoint,

wherein the virtual network path comprises a virtual wire between the a first VNIC selected from the plurality of VNICs and a second VNIC selected from the plurality of VNICs,

wherein the first VNIC is located in a first blade selected from the plurality of blades,

wherein the second VNIC is located in a second blade selected from the plurality of blades,

wherein the first blade connects to the chassis using the first PCI-E endpoint from the plurality of PCI-E endpoints and a second blade connects to the chassis using the second PCI-E endpoint from the plurality of PCI-E endpoints.

7. The blade chassis of claim 6 , wherein the control virtual machine is further configured to detect a fault where the control virtual machine is located, and wherein the control virtual machine is configured to migrate based on the fault.

8. The blade chassis of claim 7 , wherein the control virtual machine is further configured to:

checkpoint a control thread in the control virtual machine prior to migrating the control virtual machine; and

resume the control thread after migrating the control virtual machine.

9. The blade chassis of claim 6 , wherein after mapping the service processor the service processor is not visible to any of the plurality of virtual machines other than the control virtual machine.

10. A non-transitory computer readable medium comprising a plurality of executable instructions for controlling network traffic in a blade chassis, wherein the plurality of executable instructions comprises instructions to:

assign control of a network express manager located in the blade chassis to a control virtual machine selected from a plurality of virtual machines,

wherein the plurality of virtual machines is located in a plurality of blades communicatively coupled with each other via a chassis interconnect in the blade chassis,

wherein the chassis interconnect comprises a Peripheral Component Interface Express (PCI-E) backplane comprising a plurality of PCI-E endpoints,

wherein a first blade of the plurality of blades connects to the blade chassis using a first PCI-E endpoint from the plurality of PCI-E endpoints, wherein a second blade of the plurality of blades connects to the blade chassis using a second PCI-E endpoint from the plurality of PCI-E endpoints,

wherein each of the plurality of virtual machines is associated with at least one of a plurality of virtual network interface cards (VNICs),

wherein the plurality of computers share a physical network interface, and

wherein assigning control of the network express manager to the control virtual machine comprises mapping a service processor located in the network express manager as a physical device in the control virtual machine and wherein mapping the service processor is performed by a chassis processor located in the chassis;

configure the network express manager, by the control virtual machine, wherein the network express manager is configured to route network traffic in the blade chassis; and

implement a virtual network path using the network express manager, and

migrate, after implementing the virtual network path, the control virtual machine within the plurality of computers without disrupting operation of the network express manager,

wherein the virtual network path further comprises a virtual wire between a first VNIC selected from the plurality of VNICs and a second VNIC selected from the plurality of VNICs,

wherein the network express manager implements the virtual network path using a virtual switching table,

wherein the virtual switching table comprises a first mapping of a first Internet Protocol (IP) address for the first VNIC to the first PCI-E endpoint, and wherein the virtual switching table further comprises a second mapping of a second IP address for the second VNIC to the second PCI-E endpoint,

wherein the first VNIC is located in the first blade, and

wherein the second VNIC is located in the second blade.

11. The non-transitory computer readable medium of claim 10 , wherein the plurality of executable instructions further comprises instructions to:

detect a fault in the first blade, wherein instructions to migrate the control virtual machine are executed based on the fault.

12. The non-transitory computer readable medium of claim 10 , wherein the plurality of executable instructions further comprises instructions to:

checkpoint a control thread in the control virtual machine prior to migrating the control virtual machine; and

resume the control thread after migrating the control virtual machine.

13. The non-transitory computer readable medium of claim 10 , wherein instructions to assign control of the network express manager to the control virtual machine comprise instructions to map a service processor located in the network express manager as a physical device in the control virtual machine.

14. The non-transitory computer readable medium of claim 13 , wherein instructions to map the service processor are executed by a chassis processor located in the blade chassis.

15. The non-transitory computer readable medium of claim 13 , wherein after mapping the service processor the service processor is not visible to any of the plurality of virtual machines other than the control virtual machine.

Assignments (2)
MERGER AND CHANGE OF NAME Recorded Dec 16, 2015
From: ORACLE USA, INC.; SUN MICROSYSTEMS, INC.; ORACLE AMERICA, INC.
To: ORACLE AMERICA, INC.
Reel/Frame 037311/0195 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Dec 11, 2007
From: TRIPATHI, SUNAY; NORDMARK, ERIK
To: SUN MICROSYSTEMS, INC.
Reel/Frame 020229/0268 →
Continuity (1)
Related Publication 20090150883A1 · Jun 11, 2009