DISTRIBUTED NETWORK VIRTUALIZATION APPARATUS AND METHOD

• US 20130058357A1
• Filed: 07/06/2011
• Published: 03/07/2013
• Est. Priority Date: 07/06/2010
• Status: Active Grant

• Current Assignee: Nicira Incorporated (Broadcom, Inc.)
• Original Assignee: Pankaj Thakkar
• Inventors: Koponen, Teemu, Thakkar, Pankaj, Casado, Martin, Lambeth, W. Andrew
• Granted Patent: US 8,718,070 B2
• Time in Patent Office: Days
• US Class Current: 370/412
• CPC Class Codes: G06F 11/07 Responding to the occurrence of a fault, e.g. fault tolerance
• CPC Class Codes: G06F 15/17312 Routing techniques specific to parallel machines, e.g. wormhole, store and forward, shortest path problem congestion routing on a LAN H04L45/00
• CPC Class Codes: H04L 12/4633 Interconnection of networks using encapsulation techniques, e.g. tunneling
• CPC Class Codes: H04L 2101/622 Layer-2 addresses, e.g. medium access control [MAC] addresses
• CPC Class Codes: H04L 41/0816 the condition being an adaptation, e.g. in response to network events
• CPC Class Codes: H04L 41/0853 by actively collecting configuration information or by backing up configuration information
• CPC Class Codes: H04L 41/0893 Assignment of logical groups to network elements
• CPC Class Codes: H04L 41/0894 Policy-based network configuration management
• CPC Class Codes: H04L 41/0895 Configuration of virtualised networks or elements, e.g. virtualised network function or OpenFlow elements
• CPC Class Codes: H04L 41/0896 Bandwidth or capacity management, i.e. automatically increasing or decreasing capacities flow or congestion control using dynamic resource allocation, e.g. in-call renegotiation, H04L47/76
• CPC Class Codes: H04L 41/122 of virtualised topologies, e.g. software-defined networks [SDN] or network function virtualisation [NFV]
• CPC Class Codes: H04L 45/00 Routing or path finding of packets in data switching networks routing or path finding in wireless networks H04W40/00
• CPC Class Codes: H04L 45/02 Topology update or discovery
• CPC Class Codes: H04L 45/036 Updating the topology between route computation elements, e.g. between OpenFlow controllers
• CPC Class Codes: H04L 45/586 of virtual routers
• CPC Class Codes: H04L 45/76 Routing in software-defined topologies, e.g. routing between virtual machines
• CPC Class Codes: H04L 47/783 Distributed allocation of resources, e.g. bandwidth brokers
• CPC Class Codes: H04L 49/00 Packet switching elements
• CPC Class Codes: H04L 49/1546 using pipelined operation
• CPC Class Codes: H04L 49/3063 Pipelined operation
• CPC Class Codes: H04L 49/70 : Virtual switches
• CPC Class Codes: H04L 61/5007 : Internet protocol [IP] addresses

Abstract

Some embodiments provide a distributed control system for controlling managed switching elements of a network. The distributed control system comprises a first network virtualizer for converting a first set of input logical forwarding plane data to a first set of output physical control plane data. It also includes a second network virtualizer for converting a second set of input logical forwarding plane data to a second set of output physical control plane data. In some embodiments, the physical control plane data is translated into physical forwarding behaviors that direct the forwarding of data by the managed switching elements.

Claims

1. A distributed control system for controlling managed switching elements of a network, the distributed control system comprising:

a first network virtualizer for converting a first set of input logical forwarding plane data to a first set of output physical control plane data;

a second network virtualizer for converting a second set of input logical forwarding plane data to a second set of output physical control plane data; and

said physical control plane data for subsequent translation into physical forwarding behaviors that direct the forwarding of data by the managed switching elements.

2. The distributed control system of claim 1, wherein the first set of input logical forwarding plane data comprises logical forwarding data for a first set of logical data path sets, while the second set of input logical forwarding plane data comprises logical forwarding data for a second set of logical data path sets that differ from the first set of logical data path sets.

3. The distributed control system of claim 1, wherein the first virtualizer comprises:

an interface for receiving input logical forwarding plane data;

a filter for filtering out received logical forwarding plane data that is not in the first set of input logical forwarding plane data; and

a converter for converting logical forwarding plane data that is not filtered out by the filter to output physical control plane data.

4. The distributed control system of claim 3, wherein the second virtualizer comprises:

an interface for receiving input logical forwarding plane data;

a filter for filtering out received logical forwarding plane data that is not in the second set of input logical forwarding plane data; and

a converter for converting logical forwarding plane data that is not filtered out by the filter to output physical control plane data.

5. The distributed control system of claim 3,

wherein the first virtualizer is a master virtualizer of a first set of logical data paths, while the second virtualizer is a master virtualizer of a second set of logical data paths that differs from the first set of logical data paths;

wherein the filter filters out received logical forwarding plane data when the filter determines that the logical forwarding plane data does not relate to the first set of logical data paths.

6. The distributed control system of claim 3,

wherein the interface comprises a first set of tables for receiving input logical forwarding plane data;

wherein the first virtualizer'"'"'s converter comprises a table mapper that maps records in the first set of tables to records in a second set of tables;

wherein a plurality of records in the second set of tables contain the output physical control plane data; and

wherein the table mapper comprises the filter and the converter.

7. The distributed control system of claim 6, wherein the table mapper comprises:

a set of rules for specifying join operations on records of tables;

a table mapping engine for performing the join operations;

wherein when the first set of tables receives a new input logical forwarding plane data, the table mapping engine performs a set of join operations including one filtering join operation that filters out the result of the set join operations when the new input logical forwarding plane data is not in the first set of input logical forwarding plane data.

8. The distributed control system of claim 1, wherein the sets of input logical forwarding plane data are at least partially supplied by a set of controllers that translates logical control plane data to logical forwarding plane data.

9. The distributed control system of claim 1, wherein each network virtualizer comprises a monitor for supplying the virtualizer'"'"'s interface with input logical forwarding plane data by monitoring the managed switching elements.

10. The distributed control system of claim 9, wherein each monitor monitors the managed switching elements by monitoring a data storage structure that stores data regarding the managed switching elements.

11. The distributed control system of claim 1,

wherein each network virtualizer comprises a monitor for supplying the virtualizer'"'"'s interface with input logical forwarding plane data by monitoring the managed switching elements through a data storage structure that stores data regarding the managed switching elements;

wherein the virtualizer and the data storage structure operate on the same computer.

12. The network virtualizer of claim 1,

wherein each network virtualizer comprises a monitor for supplying the virtualizer'"'"'s interface with input logical forwarding plane data by monitoring the managed switching elements through a data storage structure that stores data regarding the managed switching elements;

wherein the virtualizer and the data storage structure operate on different computers.

13. The distributed control system of claim 1, wherein a set of managed switching elements perform the translation of the physical control plane data to the physical forwarding plane data.

14. A network virtualizer for converting a first set of input logical forwarding plane data to a first set of output physical control plane data, the virtualizer comprising:

an interface for receiving input logical forwarding plane data;

a filter for filtering out received logical forwarding plane data that is not in the first set of input logical forwarding plane data; and

a converter for generating physical control plane data from logical forwarding plane data that is not filtered out by the filter;

said physical control plane data for subsequent translation into physical forwarding behaviors that direct the forwarding of data by the managed switching elements.

15. The network virtualizer of claim 14,

wherein the virtualizer is a master virtualizer of a first set of logical data paths;

wherein the filter filters out received logical forwarding plane data when the filter determines that the logical forwarding plane data does not relate to the first set of logical data paths.

16. The network virtualizer of claim 14,

wherein the interface comprises a first set of tables for receiving input logical forwarding plane data;

wherein the converter comprises a table mapper that maps records in the first set of tables to records in a second set of tables;

wherein a plurality of records in the second set of tables contain the output physical control plane data; and

wherein the table mapper comprises the filter and the converter.

17. The network virtualizer of claim 16, wherein the table mapper comprises:

a set of rules for specifying join operations on records of tables;

a table mapping engine for performing the join operations;

wherein when the first set of tables receives a new input logical forwarding plane data, the table mapping engine performs a set of join operations including one filtering join operation that terminates the set join operations when the new input logical forwarding plane data is not in the first set of input logical forwarding plane data.

18. The network virtualizer of claim 14,

wherein the interface comprises a first set of tables for receiving input logical forwarding plane data and input physical control plane data;

the virtualizer further comprising;

a monitor for monitoring operation of the managed switching elements and supplying the first set of tables with physical control plane data regarding the operation of the managed switching elements;

a second set of tables for storing physical control plane data generated by the converter;

wherein upon a change of a physical control plane data record in the first set of table, the converter generates a new record that is stored in the second set of table.

19. The network virtualizer of claim 18, wherein the converter comprises a table mapper that maps records in the first set of tables to records in the second set of tables, wherein the table mapper comprises:

a set of rules for specifying join operations on records of tables;

a table mapping engine for performing the join operations;

wherein when the first set of tables receives a new input physical control plane data, the table mapping engine performs a set of join operations including one filtering join operation that terminates the set join operations when the new input physical control plane data is not associated with the first set of input logical forwarding plane data.

20. The network virtualizer of claim 14, wherein the input logical forwarding plane data are at least partially supplied by a controller that translates logical control plane data to logical forwarding plane data.

21. The network virtualizer of claim 14 further comprising a monitor for supplying the interface with input logical forwarding plane data by monitoring the managed switching elements.

22. The network virtualizer of claim 21, wherein the monitor monitors the managed switching elements by monitoring a data storage structure that stores data regarding the managed switching elements.

23. The network virtualizer of claim 14 further comprising a monitor for supplying the interface with input logical forwarding plane data by monitoring the managed switching elements through a data storage structure that stores data regarding the managed switching elements, wherein the virtualizer and the data storage structure operate on the same computer.

24. The network virtualizer of claim 14 further comprising a monitor for supplying the interface with input logical forwarding plane data by monitoring the managed switching elements through a data storage structure that stores data regarding the managed switching elements, wherein the virtualizer and the data storage structure operate on two different computers.

25. The network virtualizer of claim 14, wherein the physical control plane data comprises a plurality of logical flow entries in at least one logical forwarding table that is not accessible for input by a source of input of the logical forwarding plane data.

26. The network virtualizer of claim 25, wherein at least a plurality of the logical flow entries specify L2 or L3 forwarding rules for forwarding of data between logical MAC or IP addresses of logical managed switching elements.

27. The network virtualizer of claim 14 further comprising a publisher for storing the generated physical control plane data in a storage structure that stores data regarding the operation of the managed switching elements.

28. For a distributed system that has a plurality of virtualizer applications for constructing a plurality of logical data path sets, a machine readable medium for storing a program for mapping one set of logical forwarding plane data to a set of physical control plane data, the program comprising sets of instructions for:

receiving an input logical forwarding plane data for a particular logical data path set;

determining whether the particular virtualizer application is responsible for the particular logical data path set;

when the particular virtualizer application is not responsible for the particular logical data path set, discarding the logical forwarding plane data; and

when the particular virtualizer application is responsible for the particular logical data path set, generating physical control plane data from the logical forwarding plane data, wherein the input logical forwarding plane data is a smaller subset of the physical control plane data.

29. A distributed system including a plurality of virtualizer applications for constructing a plurality of logical data path sets, a particular virtualizer application of the distributed system comprising:

a first module for receiving an input logical forwarding plane data for a particular logical data path set;

a second module for determining whether the particular virtualizer application is responsible for the particular logical data path set; and

a third module for;

discarding the logical forwarding plane data when the particular virtualizer application is not responsible for the particular logical data path set; and

generating physical control plane data from the logical forwarding plane data when the particular virtualizer application is responsible for the particular logical data path set, wherein the input logical forwarding plane data is a smaller subset of the physical control plane data.