Apparatus, system, and method for improving convergence across virtual control and forwarding planes within network devices

• US 10,645,000 B1
• Filed: 06/03/2018
• Issued: 05/05/2020
• Est. Priority Date: 06/03/2018
• Status: Active Grant

• Current Assignee: Juniper Networks Incorporated
• Original Assignee: Juniper Networks Incorporated
• Inventors: Nayak, Manoj, Khan, Tabrez Ahmed, Puttur, Rafik
• Primary Examiner(s): Marcelo, Melvin C
• Application Number:
  US15/996,517
• Time in Patent Office: 702 Days
• CPC Class Codes: H04L 45/586 of virtual routers
• CPC Class Codes: H04L 45/64 using an overlay routing layer

Abstract

A disclosed method may include (1) identifying a first virtual forwarding plane of a physical network device that connects to a virtual control plane of the physical network device, (2) obtaining, by the first virtual forwarding plane, a set of data objects from the virtual control plane of the network device, (3) identifying a second virtual forwarding plane of the physical network device that connects to the first virtual forwarding plane of the physical network device, and then (4) providing, by the first virtual forwarding plane, at least a portion of the set of data objects to the second virtual forwarding plane via a shared memory segment that is accessible to both the first virtual forwarding plane and the second virtual forwarding plane. Various other systems and methods are also disclosed.

Claims

1. A method comprising:

identifying a first virtual forwarding plane of a physical network device that connects to a virtual control plane of the physical network device;

obtaining, by the first virtual forwarding plane, a set of data objects from the virtual control plane of the physical network device;

identifying a second virtual forwarding plane of the physical network device that connects to the first virtual forwarding plane of the physical network device; and

providing, by the first virtual forwarding plane, at least a portion of the set of data objects to the second virtual forwarding plane via a shared memory segment that is accessible to both the first virtual forwarding plane and the second virtual forwarding plane.

2. The method of claim 1, further comprising obtaining, by the second virtual forwarding plane, at least one data object that is excluded from the set of data objects from the virtual control plane via a network communication protocol.

3. The method of claim 1, further comprising:

obtaining, by the second virtual forwarding plane, at least one data object included in the set of data objects from the virtual control plane via a network communication protocol;

determining that the second virtual forwarding plane has already obtained the data object from the first virtual forwarding plane via the shared memory segment; and

in response to the determination, dropping the data object from the second virtual forwarding plane.

4. The method of claim 1, further comprising:

processing the set of data objects at the first virtual forwarding plane;

during the processing of the set of data objects;

identifying a first subset of the data objects as belonging to the first virtual forwarding plane; and

marking the first subset of data objects to indicate which of the data objects belong to the first virtual forwarding plane; and

wherein providing the portion of the set of data objects to the second virtual forwarding plane comprises;

identifying a second subset of the data objects that do not belong to the first virtual forwarding plane; and

providing the second subset of data objects to the second virtual forwarding plane via the shared memory segment.

5. The method of claim 4, wherein providing the second subset of data objects to the second virtual forwarding plane comprising rejecting, by the second virtual forwarding plane, any data objects marked as belonging to the first virtual forwarding plane.

6. The method of claim 4, further comprising:

calculating an amount of memory consumed by the first set of data objects at the first virtual forwarding plane;

computing a memory limitation that facilitates limiting an amount of memory allocated by the second virtual forwarding plane to the second subset of data objects by subtracting the amount of memory consumed by the first set of data objects from a total amount of heap memory for the first virtual forwarding plane; and

applying the memory limitation to the second virtual forwarding plane.

7. The method of claim 1, further comprising:

identifying a third virtual forwarding plane of the physical network device that connects to the second virtual forwarding plane of the physical network device; and

providing, by the second virtual forwarding plane, at least some of the set of data objects to the third virtual forwarding plane via the shared memory segment.

8. The method of claim 7, further comprising:

processing the portion of the set of data objects at the second virtual forwarding plane;

during the processing of the portion of the set of data objects;

identifying a first subset of the data objects as belonging to the second virtual forwarding plane; and

marking the first subset of the data objects to indicate which of the data objects belong to the second virtual forwarding plane; and

wherein providing the at least some of the set of data objects to the third virtual forwarding plane comprises;

identifying a second subset of the data objects that do not belong to the second virtual forwarding plane; and

providing the second subset of data objects to the third virtual forwarding plane via the shared memory segment.

9. The method of claim 8, further comprising obtaining, by the third virtual forwarding plane, at least one data object that is excluded from the second subset of data objects from the virtual control plane.

10. The method of claim 9, further comprising:

obtaining, by the third virtual forwarding plane, at least one data object included in the second set of data objects from the virtual control plane;

determining that the second virtual forwarding plane has already obtained the data object from the second virtual forwarding plane via the shared memory segment; and

in response to the determination, dropping the data object from the third virtual forwarding plane.

11. The method of claim 8, further comprising:

calculating an amount of memory consumed by the first subset of data objects at the second virtual forwarding plane;

computing a memory limitation that facilitates limiting an amount of memory allocated by the third virtual forwarding plane to the second subset of data objects by subtracting the amount of memory consumed by the first set of data objects from a total amount of heap memory for the second virtual forwarding plane; and

applying the memory limitation to the third virtual forwarding plane.

12. The method of claim 1, wherein:

the first virtual forwarding plane operates on a first flexible physical interface card concentrator with;

a first amount of processing power; and

a first memory size; and

the second virtual forwarding plane operates on a second flexible physical interface card concentrator with;

a second amount of processing power that is less than the first amount of processing power; and

a second memory size that is smaller than the first memory size.

13. The method of claim 12, the shared memory segment comprises random-access memory included on the first flexible physical interface card concentrator.

14. An apparatus comprising:

a routing engine that;

deploys a virtual control plane of a network device; and

stores a set of data objects in connection with the virtual control plane;

a first flexible physical interface card concentrator that;

deploys a first portion of a virtual forwarding plane of the network device; and

obtains the set of data objects from the routing engine in connection with the virtual forwarding plane;

a second flexible physical interface card concentrator that;

deploys a second portion of the virtual forwarding plane of the network device; and

obtains, in connection with the virtual forwarding plane, at least a portion of the set of data objects from the first flexible physical interface card concentrator via a shared memory segment that is accessible to both the first flexible physical interface card concentrator and the second flexible physical interface card concentrator.

15. The apparatus of claim 14, wherein the second flexible physical interface card concentrator obtains, in connection with the second portion of the virtual forwarding plane, at least one data object that is excluded from the set of data objects from the virtual control plane.

16. The apparatus of claim 14, wherein the second flexible physical interface card concentrator:

obtains, in connection with the second portion of the virtual forwarding plane, at least one data object included in the set of data objects from the routing engine;

determines that the second portion of the virtual forwarding plane has already obtained the data object from the first flexible physical interface card concentrator via the shared memory segment; and

drops, in response to the determination, the data object from the second portion of the virtual forwarding plane.

17. The apparatus of claim 14, wherein:

the first flexible physical interface card concentrator;

processes the set of data objects;

during the processing of the set of data objects;

identifies a first subset of the data objects as belonging to the first portion of the virtual forwarding plane;

marks the first subset of data objects to indicate which of the data objects belong to the first portion of the virtual forwarding plane; and

identifies a second subset of the data objects that do not belong to the first portion of the virtual forwarding plane; and

provides the second subset of data objects to the second portion of the virtual forwarding plane via the shared memory segment.

18. The apparatus of claim 17, wherein the second flexible physical interface card concentrator rejects any data objects marked as belonging to the first portion of the virtual forwarding plane.

19. The apparatus of claim 17, wherein:

the first flexible physical interface card concentrator;

calculates an amount of memory consumed by the first set of data objects at the first portion of the virtual forwarding plane; and

notifies the routing engine of the amount of memory consumed by the first set of data objects at the first portion of the virtual forwarding plane; and

the routing engine;

computes a memory limitation that facilitates limiting an amount of memory allocated by the second portion of the virtual forwarding plane to the second subset of data objects by subtracting the amount of memory consumed by the first set of data objects from a total amount of heap memory for the first portion of the virtual forwarding plane; and

applies the memory limitation to the second portion of the virtual forwarding plane.

20. A network device comprising:

a routing engine that;

deploys a virtual control plane of the network device; and

stores a set of data objects in connection with the virtual control plane;

a first flexible physical interface card concentrator that;

deploys a first portion of a virtual forwarding plane of the network device; and

obtains the set of data objects from the routing engine in connection with the virtual forwarding plane;

a second flexible physical interface card concentrator that;

deploys a second portion of the virtual forwarding plane of the network device; and

obtains, in connection with the virtual forwarding plane, at least a portion of the set of data objects from the first flexible physical interface card concentrator via a shared memory segment that is accessible to both the first flexible physical interface card concentrator and the second flexible physical interface card concentrator.