Fibre channel over ethernet

• US 8,238,347 B2
• Filed: 04/06/2006
• Issued: 08/07/2012
• Est. Priority Date: 10/22/2004
• Status: Active Grant

• Current Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Original Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Inventors: DeSanti, Claudio, Gai, Silvano, Baldini, Andrea
• Primary Examiner(s): Sheikh, Ayaz
• Assistant Examiner(s): AGA, SORI A
• Application Number:
  US11/400,671
• Publication Number:
  US 20060251067A1
• Time in Patent Office: 2,315 Days
• Field of Search: None
• US Class Current: 370/395.5
• CPC Class Codes: H04L 12/462 LAN interconnection over a bridge based backbone
• CPC Class Codes: H04L 12/4633 Interconnection of networks using encapsulation techniques, e.g. tunneling
• CPC Class Codes: H04L 2101/604 Address structures or formats
• CPC Class Codes: H04L 2101/622 Layer-2 addresses, e.g. medium access control [MAC] addresses
• CPC Class Codes: H04L 2212/00 Encapsulation of packets
• CPC Class Codes: H04L 49/351 for local area network [LAN], e.g. Ethernet switches
• CPC Class Codes: H04L 49/357 Fibre channel switches

Abstract

A Data Center Ethernet (“DCE”) network and related methods and device are provided. A DCE network simplifies the connectivity of data centers and provides a high bandwidth, low latency network for carrying Ethernet, storage and other traffic. A Fiber Channel (“FC”) frame, including FC addressing information, is encapsulated in an Ethernet frame for transmission on a Data Center Ethernet (“DCE”) network. The Ethernet address fields may indicate that the frame includes an FC frame, e.g., by the use of a predetermined Organization Unique Identifier (“OUI”) code in the D_MAC field, but also include Ethernet addressing information. Accordingly, the encapsulated frames can be forwarded properly by switches in the DCE network whether or not these switches are configured for operation according to the FC protocol. Accordingly, only a subset of Ethernet switches in the DCE network needs to be FC-enabled. Only switches so configured will require an FC Domain_ID.

Claims

1. A network, comprising:

a first plurality of switches configured for operation according to the Fibre Channel (“

FC”

) protocol; and

a second plurality of switches configured for operation according to the Ethernet protocol but not configured for operation according to the FC protocol, wherein each of the first plurality of switches and each of the second plurality of switches is configured to simultaneously carry traffic according to multiple protocols on multiple virtual lanes of the same physical link, each one of the multiple protocols being assigned to a different set of one or more of the multiple virtual lanes, wherein each different set of one or more of the multiple virtual lanes carries traffic according to the one of the multiple protocols that has been assigned;

wherein the multiple protocols include a first protocol with drop behavior and a second protocol with no-drop behavior, and wherein a first one of the multiple virtual lanes of the physical link is assigned as a drop lane such that the first one of the multiple virtual lanes carries traffic in the first protocol with drop behavior, and a second one of the multiple virtual lanes of the physical link is assigned as a no-drop lane such that the second one of the multiple virtual lanes carries traffic in the second protocol with no-drop behavior.

2. The network of claim 1, wherein at least some of the traffic on the network comprises FC frames that are encapsulated as Ethernet frames.

3. The network of claim 1, wherein each switch of the first plurality of switches is assigned an FC Domain_ID and wherein each switch of the second plurality of switches is not assigned an FC Domain_ID.

4. The network of claim 1, wherein each of the second plurality of switches is configured to forward any encapsulated FC frame to a predetermined switch of the first plurality of switches.

5. The network of claim 1, wherein the multiple protocols include at least the Ethernet protocol and the FC protocol.

6. The network of claim 1, wherein the multiple protocols include a clustering protocol.

7. The network of claim 2, wherein each of the Ethernet frames has an Ethertype field that indicates the Ethernet frame encapsulates an FC frame.

8. The network of claim 2, wherein each device in the network that is configured to encapsulate an FC frame in an Ethernet frame is also configured to perform the following steps:

determine whether the Ethernet frame has a minimum size required by an Ethernet protocol; and

pad the Ethernet frame when it is determined that the Ethernet frame does not have a minimum size required by the Ethernet protocol.

9. The network of claim 4, wherein any FC frame is encapsulated in an Ethernet frame.

10. The network of claim 9, further comprising a third plurality of host devices, wherein an FC frame is formed and encapsulated in an Ethernet frame by one of a host device of third plurality of host devices or a switch of the first plurality of switches.

11. The network of claim 10, wherein the Ethernet frame comprises:

a destination media access control address (“

D_MAC”

) field having a first portion with an Organization Unique Identifier (“

OUI”

) value indicating that the D_MAC address is mapping an FC destination identification (“

FC D_ID”

) and a second portion that is the mapped FC D_ID information corresponding to the destination identification (“

D_ID”

) of the encapsulated FC frame; and

a source MAC (S_MAC) field that indicates the MAC address of the device that generated the FC frame.

12. The network of claim 11, wherein the predetermined switch of the first plurality of switches is configured to perform the following steps:

receive the Ethernet frame;

determine that the Ethernet frame contains an FC frame;

decapsulate the FC frame;

either forward or locally process the decapsulated FC frame.

13. The network of claim 12, wherein the decapsulated FC frame is locally processed and wherein the processing implies a response, and wherein the predetermined switch is further configured to perform the following steps:

form an appropriate response;

encapsulate the response in an additional Ethernet frame having a D_MAC field indicating the device that sent the received frame; and

transmit the additional Ethernet frame to the device that sent the received frame.

14. The network of claim 12, wherein the determining step is based on at least one of the Ethertype value and the information in the D_MAC field.

15. The network of claim 13, wherein the predetermined switch is further configured to perform the following steps:

determine whether the additional Ethernet frame has a minimum size required by an Ethernet protocol; and

pad the additional Ethernet frame when it is determined that the additional Ethernet frame does not have a minimum size required by the Ethernet protocol.

26. The network of claim 1 wherein the first protocol is Ethernet, the second protocol is Fibre Channel, the first one of the multiple virtual lanes of the physical link is assigned as an Ethernet lane to carry Ethernet traffic, and the second one of the multiple virtual lanes of the physical link is assigned as a Fibre Channel lane to carry Fibre Channel traffic.

27. The network of claim 1, wherein the same physical link is dynamically partitioned into the multiple virtual lanes.

28. The network of claim 27, wherein dynamic partitioning is performed by adding an extension header to traffic carried on the link.

29. The network of claim 27, wherein dynamic partitioning is performed using one or more fields of a header of traffic carried on the link.

16. A gateway comprising:

means for receiving a first Fibre Channel (“

FC”

) frame from a storage network;

means for encapsulating the first FC frame as a first Ethernet frame; and

means for transmitting the first Ethernet frame on a virtual lane of a Data Center Ethernet (DCE) network, the DCE network comprising switches configured to simultaneously carry traffic according to multiple protocols on multiple virtual lanes of the same physical link, each one of the multiple protocols being assigned to a different set of one or more of the multiple virtual lanes, wherein each different set of one or more of the multiple virtual lanes carries traffic according to the one of the multiple protocols that has been assigned;

wherein the multiple protocols include a first protocol with drop behavior and a second protocol with no-drop behavior, and wherein a first one of the multiple virtual lanes of the physical link is assigned as a drop lane such that the first one of the multiple virtual lanes carries traffic in the first protocol with drop behavior, and a second one of the multiple virtual lanes of the physical link is assigned as a no-drop lane such that the second one of the multiple virtual lanes carries traffic in the second protocol with no-drop behavior.

17. The gateway of claim 16, further comprising:

means for receiving a second Ethernet frame comprising an encapsulated second FC frame;

means for decapsulating the second Ethernet frame; and

means for transmitting the second FC frame to the storage network.

30. The gateway as recited in claim 16, wherein each of the multiple virtual lanes has a separate set of credits.

31. The gateway as recited in claim 16, wherein each of the multiple virtual lanes is treated independently from one another with respect to scheduling, flow control, and resource assignment.

32. The gateway as recited in claim 16, wherein each of the multiple virtual lanes is assigned a different guaranteed bandwidth.

33. The gateway as recited in claim 16, further comprising:

means for providing virtual lane information pertaining to behavior of each of the multiple virtual lanes.

34. The gateway as recited in claim 16, further comprising:

means for determining whether a host supports Data Center Ethernet (DCE); and

means for providing virtual lane information pertaining to behavior of each of the multiple virtual lanes to the host upon determining that the host supports DCE.

18. A gateway comprising:

a plurality of Fibre Channel (“

FC”

) ports configured for communication with a FC network; and

a plurality of Ethernet ports configured for communication with a Data Center Ethernet (DCE) network, the DCE network comprising switches configured to simultaneously carry traffic according to multiple protocols on multiple virtual lanes of the same physical link, each one of the multiple protocols being assigned to a different set of one or more of the multiple virtual lanes, wherein each different set of one or more of the multiple virtual lanes carries traffic according to the one of the multiple protocols that has been assigned;

wherein the multiple protocols include a first protocol with drop behavior and a second protocol with no-drop behavior, and wherein a first one of the multiple virtual lanes of the physical link is assigned as a drop lane such that the first one of the multiple virtual lanes carries traffic in the first protocol with drop behavior, and a second one of the multiple virtual lanes of the physical link is assigned as a no-drop lane such that the second one of the multiple virtual lanes carries traffic in the second protocol with no-drop behavior.

19. The gateway as recited in claim 18, further comprising:

at least one logic device configured to perform the following steps;

receive a first FC frame from one of the plurality of FC ports;

encapsulate the first FC frame as a first Ethernet frame; and

forward the Ethernet frame to one of the plurality of Ethernet ports for transmission on a virtual lane of the DCE network.

35. The gateway of claim 19, wherein at least one logic device is configured to perform the following steps:

receive a second Ethernet frame from one of the plurality of Ethernet FC ports, the second Ethernet frame encapsulating a second FC frame;

decapsulate the second Ethernet frame; and

forward the second FC frame to one of the plurality of FC ports for transmission on the storage network.

36. The gateway as recited in claim 19, wherein the encapsulation step comprises:

inserting a MAC address of a first device performing the forming and encapsulating steps into a source MAC (“

S_MAC”

) field of the first Ethernet frame;

inserting a code in an Ethertype field of the first Ethernet frame, the code indicating that the first Ethernet frame includes the first FC frame; and

including FC addressing information with other encapsulated fields of the FC frame.

20. A network, comprising:

a first plurality of switches configured for operation according to the Fibre Channel (“

FC”

) protocol; and

a second plurality of switches configured to simultaneously carry traffic according to multiple protocols on multiple virtual lanes of the same physical link, the second plurality of switches comprising;

a first portion of switches that are Fibre Channel over Ethernet (“

FCoE”

) switches configured to process FCoE frames, which are FC frames encapsulated in Ethernet frames; and

a second portion of switches that are not FCoE switches, each one of the multiple protocols being assigned to a different set of one or more of the multiple virtual lanes, wherein each different set of one or more of the multiple virtual lanes carries traffic according to only the one of the multiple protocols that has been assigned;

wherein the multiple protocols include a first protocol with drop behavior and a second protocol with no-drop behavior, and wherein a first one of the multiple virtual lanes of the physical link is assigned as a drop lane such that the first one of the multiple virtual lanes carries traffic in the first protocol with drop behavior, and a second one of the multiple virtual lanes of the physical link is assigned as a no-drop lane such that the second one of the multiple virtual lanes carries traffic in the second protocol with no-drop behavior.

21. The network of claim 20, wherein the FCoE switches are configured as logical neighbors by E_Port to E_Port operation through MAC-in-MAC tunnels.

22. The network of claim 20, wherein the FCoE switches that are topologically in the network'"'"'s core are configured to be at logical edges of the network.

23. The network of claim 20, wherein at least some of the first portion of switches are configured to encapsulate FC frames as Ethernet frames, thereby forming FCoE frames, and further configured to decapsulate FCoE frames and extract the FC frames.

24. The network of claim 21, wherein only a MAC-in-MAC destination address is used to forward packets between logical neighbors.

25. The network of claim 22, wherein FCoE encapsulation is performed only by switches configured to be at logical edges of the network.