Virtual cluster switching

Abstract

One embodiment of the present invention provides a switch system. The switch includes one or more ports on the switch configured to transmit packets encapsulated based on a first protocol. The switch further includes a control mechanism. During operation, the control mechanism forms a logical switch based on a second protocol, receives an automatically assigned identifier for the logical switch without requiring manual configuration of the identifier, and joins a virtual cluster switch.

1 - 20 . (canceled) 
     
     
         21 . A switch, comprising:
 a control module configured to:
 provide a control plane with configuration capabilities; 
 join a network of interconnected switches via the control plane to become a member switch of the network of interconnected switches; and 
 control the network of interconnected switches as a single logical switch based on the control plane; and 
   a traffic management module configured to:
 in response to determining congestion in a local queue, extract a source media access control (MAC) address of a packet in the local queue, wherein the source MAC address is encapsulated in an encapsulation header, which is used to transport packets among member switches within the network of interconnected; and 
 construct a congestion notification frame for a device corresponding to the extracted source MAC address. 
   
     
     
         22 . The switch of  claim 21 ,
 wherein the traffic management module is further configured to associate incoming traffic with two outgoing ports corresponding to two respective equal-cost data paths to a common destination, thereby facilitating load balancing.   
     
     
         23 . The switch of  claim 21 ,
 wherein the traffic management module is further configured to perform priority-based flow control on a respective link.   
     
     
         24 . The switch of  claim 21 ,
 wherein the traffic management module is further configured to perform enhanced transmission selection for ingress traffic; and   wherein the enhanced transmission selection comprises enforcement of bandwidth allocation to a first class of traffic.   
     
     
         25 . The switch of  claim 21 , wherein the traffic management module is further configured to, in response to a first class of traffic not using entire allocated bandwidth, allocate the unused bandwidth to a second class of traffic. 
     
     
         26 . The switch of  claim 21 , further comprising a number of links grouped into a hybrid trunk, wherein the hybrid trunk includes one or more trunk groups. 
     
     
         27 . The switch of  claim 21 ,
 wherein the one or more physical switches are allowed to be coupled in an arbitrary topology.   
     
     
         28 . The switch of  claim 21 , wherein a packet received at the switch is encapsulated based on a protocol other than Ethernet. 
     
     
         29 . The switch of  claim 28 , wherein the protocol is one or more of:
 a Transparent Interconnection of Lots of Links (TRILL) protocol, wherein the packets are encapsulated in TRILL headers;   an Internet Protocol (IP), wherein the packets are encapsulated in IP headers; and   a multi-protocol label switching (MPLS) protocol, wherein the packets are encapsulated in MPLS headers.   
     
     
         30 . The switch of  claim 21 , wherein a common identifier is associated with the network of interconnected switches. 
     
     
         31 . A method, comprising:
 providing a control plane with automatic configuration capabilities;   joining, by a switch, a network of interconnected switches via the control plane to become a member switch of the network of interconnected switches;   controlling the network of interconnected switches as a single logical switch based on the control plane;   in response to determining congestion in a local queue of the switch, extracting a source media access control (MAC) address of a packet in the local queue, wherein the source MAC address is encapsulated in an encapsulation header, which is used to transport packets among member switches within the network of interconnected switches; and   constructing a congestion notification frame for a device corresponding to the extracted source MAC address.   
     
     
         32 . The method of  claim 31 , further comprising associating incoming traffic with two outgoing ports corresponding to two respective equal-cost data paths to a common destination, thereby facilitating load balancing. 
     
     
         33 . The method of  claim 31 , further comprising performing enhanced transmission selection for ingress traffic;
 wherein the enhanced transmission selection comprises enforcement of bandwidth allocation to a first class of traffic.   
     
     
         34 . The method of  claim 31 , further comprising performing priority-based flow control on a respective link. 
     
     
         35 . The method of  claim 31 , further comprising, in response to a first class of traffic not using entire allocated bandwidth, allocating the unused bandwidth to a second class of traffic. 
     
     
         36 . The method of  claim 31 , further comprising grouping a number of links into a hybrid trunk, wherein the hybrid trunk includes one or more trunk groups. 
     
     
         37 . A switch comprising:
 a control module comprising:
 a control plane means for providing a control plane with configuration capabilities; 
 a joining means for joining a network of interconnected switches via the control plane to become a member switch of the network of interconnected switches; 
 an control means for controlling the network of interconnected switches as a single logical switch based on the control plane; and 
 a traffic management means for, in response to determining congestion in a local queue of the switch, extracting a source media access control (MAC) address of a packet in the local queue, wherein the source MAC address is encapsulated in an encapsulation header, which is used to transport packets among member within the network of interconnected switches; and 
 a notification means for constructing a congestion notification frame for a device corresponding to the extracted source MAC address.