Ethernet extension for the data center

• US 8,565,231 B2
• Filed: 05/20/2011
• Issued: 10/22/2013
• 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: Gai, Silvano, Edsall, Thomas, Cafiero, Luca, Bergamasco, Davide, Dutt, Dinesh, Bonomi, Flavio
• Primary Examiner(s): Ngo, Ricky
• Assistant Examiner(s): SAMUEL, DEWANDA A
• Application Number:
  US13/112,824
• Publication Number:
  US 20110222402A1
• Time in Patent Office: 886 Days
• Field of Search: 370/412, 370/466, 370/389, 370/241, 370/252, 370/400
• US Class Current: 370/389
• CPC Class Codes: H04L 12/4633 Interconnection of networks using encapsulation techniques, e.g. tunneling
• CPC Class Codes: H04L 12/4641 Virtual LANs, VLANs, e.g. virtual private networks [VPN] LAN interconnection over a bridge based backbone H04L12/462; encapsulation techniques H04L12/4633; routing of packets H04L45/00; packet switches H04L49/00; virtual private networks for security H04L63/0272
• CPC Class Codes: H04L 47/10 Flow control; Congestion control
• CPC Class Codes: H04L 47/13 in a LAN segment, e.g. ring or bus
• CPC Class Codes: H04L 47/24 Traffic characterised by specific attributes, e.g. priority or QoS
• CPC Class Codes: H04L 47/266 Stopping or restarting the source, e.g. X-on or X-off
• CPC Class Codes: H04L 47/32 by discarding or delaying data units, e.g. packets or frames
• CPC Class Codes: H04L 47/39 Credit based
• CPC Class Codes: H04L 69/22 Parsing or analysis of headers

Abstract

The present invention provides methods and devices for implementing a Low Latency Ethernet (“LLE”) solution, also referred to herein as a Data Center Ethernet (“DCE”) solution, which simplifies the connectivity of data centers and provides a high bandwidth, low latency network for carrying Ethernet and storage traffic. Some aspects of the invention involve transforming FC frames into a format suitable for transport on an Ethernet. Some preferred implementations of the invention implement multiple virtual lanes (“VLs”) in a single physical connection of a data center or similar network. Some VLs are “drop” VLs, with Ethernet-like behavior, and others are “no-drop” lanes with FC-like behavior. Some preferred implementations of the invention provide guaranteed bandwidth based on credits and VL. Active buffer management allows for both high reliability and low latency while using small frame buffers. Preferably, the rules for active buffer management are different for drop and no drop VLs.

Claims

1. A method for carrying more than one type of traffic on a single physical link, the method comprising:

logically partitioning traffic on a physical link of a network device into a plurality of virtual lanes;

receiving first traffic into the physical link of the network device;

determining whether the first traffic indicates a first virtual lane based on the first traffic indicating a protocol that recovers from dropping of the first traffic or indicates a second virtual lane based on the first traffic indicating a protocol that does not recover from dropping of the first traffic; and

applying a first set of rules or a second set of rules to the first traffic based on whether it is determined that the first traffic indicates the first lane or the second lane, wherein the first set of rules applies a probabilistic drop function to the first traffic in response to latency and the second set of rules does not cause frames from the second traffic to be dropped in response to latency.

2. The method of claim 1, further comprising the step of differentiating service per virtual lane.

3. The method of claim 1, wherein it is determined that the first traffic indicates a Fibre Channel protocol that does not recover from dropping of the first traffic and based on such determination, the second set of rules are applied to the first traffic so as to not cause frames to be dropped from the first traffic in response to latency.

4. The method of claim 1, wherein it is determined that the first traffic indicates an Ethernet protocol that recovers from dropping of the first traffic and based on such determination, the first set of rules are applied to the first traffic so as to apply a probabilistic drop function to the first traffic in response to deadlocks.

5. The method of claim 1, further comprising the step of implementing flow control per virtual lane by using one or more of a buffer-to-buffer crediting scheme and PAUSE frames.

6. The method of claim 5, wherein the buffer-to-buffer crediting scheme comprises crediting according to one of frame size or a number of frames.

7. The method of claim 5, wherein buffer-to-buffer credits are indicated via one of a data frame or a control frame.

8. A network device, comprising:

one or more processors;

one or more memory, wherein at least one of the processors and memory are adapted for;

logically partitioning traffic on a physical link of a network device into a plurality of virtual lanes;

receiving first traffic into the physical link of the network device;

determining whether the first traffic indicates a first virtual lane based on the first traffic indicating a protocol that recovers from dropping of the first traffic or indicates a second virtual lane based on the first traffic indicating a protocol that does not recover from dropping of the first traffic; and

applying a first set of rules or a second set of rules to the first traffic based on whether it is determined that the first traffic indicates the first lane or the second lane, wherein the first set of rules applies a probabilistic drop function to the first traffic in response to latency and the second set of rules does not cause frames from the second traffic to be dropped in response to latency.

9. The network device of claim 8, wherein the at least one of the processors and memory are further adapted for differentiating service per virtual lane.

10. The network device of claim 8, wherein it is determined that the first traffic indicates a Fibre Channel protocol that does not recover from dropping of the first traffic and based on such determination, the second set of rules are applied to the first traffic so as to not cause frames to be dropped from the first traffic in response to latency.

11. The network device of claim 8, wherein it is determined that the first traffic indicates an Ethernet protocol that recovers from dropping of the first traffic and based on such determination, the first set of rules are applied to the first traffic so as to apply a probabilistic drop function to the first traffic in response to deadlocks.

12. The network device of claim 8, wherein the at least one of the processors and memory are further adapted for implementing flow control per virtual lane by using one or more of a buffer-to-buffer crediting scheme and PAUSE frames.

13. The network device of claim 12, wherein the buffer-to-buffer crediting scheme comprises crediting according to one of frame size or a number of frames.

14. A non-transitory machine readable storage medium and having computer program instructions stored thereon that are arranged to perform the following operations:

logically partitioning traffic on a physical link of a network device into a plurality of virtual lanes;

receiving first traffic into the physical link of the network device;

determining whether the first traffic indicates a first virtual lane based on the first traffic indicating a protocol that recovers from dropping of the first traffic or indicates a second virtual lane based on the first traffic indicating a protocol that does not recover from dropping of the first traffic; and

applying a first set of rules or a second set of rules to the first traffic based on whether it is determined that the first traffic indicates the first lane or the second lane, wherein the first set of rules applies a probabilistic drop function to the first traffic in response to latency and the second set of rules does not cause frames from the second traffic to be dropped in response to latency.

15. The non-transitory machine readable storage medium of claim 14, wherein the computer program instructions are further arranged to differentiate service per virtual lane.

16. The non-transitory machine readable storage medium of claim 14, wherein it is determined that the first traffic indicates a Fibre Channel protocol that does not recover from dropping of the first traffic and based on such determination, the second set of rules are applied to the first traffic so as to not cause frames to be dropped from the first traffic in response to latency.

17. The non-transitory machine readable storage medium of claim 14, wherein it is determined that the first traffic indicates an Ethernet protocol that recovers from dropping of the first traffic and based on such determination, the first set of rules are applied to the first traffic so as to apply a probabilistic drop function to the first traffic in response to deadlocks.

18. The non-transitory machine readable storage medium of claim 14, wherein the computer program instructions are further arranged to implement flow control per virtual lane by using one or more of a buffer-to-buffer crediting scheme and PAUSE frames.

19. The non-transitory machine readable storage medium of claim 18, wherein the buffer-to-buffer crediting scheme comprises crediting according to one of frame size and a number of frames.

20. The non-transitory machine readable storage medium of claim 18, wherein buffer-to-buffer credits are indicated via one of a data frame or a control frame.