Single-failure protection in a load-balanced network architecture
View Patent ↗Techniques for single-failure protection in load-balanced network architectures are disclosed. For example, in one aspect of the invention, a technique for processing a traffic flow in a communication network comprising a plurality of nodes, the traffic flow being deliverable from a source node to at least one destination node via one or more intermediate nodes, comprises the following steps/operations. The traffic flow is split at the source node into a plurality of parts. The parts are distributed to respective ones of the intermediate nodes such that the parts are routed from the source node to the at least one destination node in a disjoint manner.
1. A method of processing a traffic flow in a communication network comprising a plurality of nodes, the traffic flow being deliverable from a source node to at least one destination node via one or more intermediate routing nodes, the method comprising the steps of:
splitting the traffic flow at the source node into a plurality of N parts; and
distributing the N parts such that the N parts are first routed from the source node to respective ones of N intermediate routing nodes in a disjoint manner and then subsequently routed from the respective ones of the N intermediate routing nodes to the at least one destination node in a disjoint manner, such that no two paths between the source node and the at least one destination node have any nodes or links in common to prevent a single link failure or a single node failure from affecting the complete traffic flow between any source-destination node pair,
wherein a first routing step of the distributing step comprises finding a first set of disjoint paths from the source node to the respective ones of the N intermediate routing nodes, and
wherein a subsequent routing step of the distributing step comprises finding a second set of disjoint paths from each of the respective ones of the N intermediate routing nodes to the at least one destination node.
2. The method of claim 1 , wherein the distributing step further comprises the step of determining one or more disjoint paths from the source node to at least a portion of the N intermediate routing nodes.
3. The method of claim 1 , wherein the distributing step further comprises the step of determining one or more disjoint paths from at least a portion of the N intermediate routing nodes to the at least one destination node.
4. The method of claim 1 , wherein the subsequent routing step is performed by the respective ones of the N intermediate routing nodes.
5. A method of processing a traffic flow in a communication network comprising a plurality of nodes, the traffic flow being deliverable from a source node to at least one destination node via one or more intermediate routing nodes, the method comprising the steps of:
splitting the traffic flow at the source node into a plurality of N parts; and
distributing the N parts to respective ones of N intermediate routing nodes such that the N parts are routed from the source node to the at least one destination node in a disjoint manner such that no two paths between the source node and the at least one destination node have any nodes or links in common to prevent a single link failure or a single node failure from affecting the complete traffic flow between any source-destination node pair;
wherein the distributing step further comprises the step of distributing, at most, two parts of the traffic flow on any given link of the network.
6. The method of claim 1 , wherein distributing step further comprises distributing the N parts of the traffic flow to the respective ones of the N intermediate routing nodes over pre-provisioned circuits each configured to support a corresponding one of the N parts.
7. The method of claim 1 , wherein each of the N parts follows one of N different paths from the source node to the destination node via a corresponding one of the N intermediate routing nodes.
8. A method of processing a traffic flow in a communication network comprising a plurality of nodes, the traffic flow being deliverable from a source node to at least one destination node via one or more intermediate routing nodes, the method comprising the steps of:
introducing two parity packets for every k−2 packets of traffic at the source node, where k is the total number of packets;
splitting the traffic flow with the parity packets at the source node into a plurality of k parts; and
distributing the k parts to respective ones of the intermediate routing nodes such that the k parts are routed from the source node to the at least one destination node in a disjoint manner where no two paths between the source node and the at least one destination node have any nodes or links in common to prevent a single link failure or a single node failure from affecting the complete traffic flow between any source-destination node pair.
9. Apparatus for use in processing a traffic flow in a communication network comprising a plurality of nodes, the traffic flow being deliverable from a source node to at least one destination node via one or more intermediate routing nodes, the apparatus comprising:
a memory; and
a processor coupled to the memory, the processor being operative to split the traffic flow into a plurality of N parts, and distribute the N parts such that the N parts are first routed from the source node to respective ones of N intermediate routing nodes in a disjoint manner and then subsequently routed from the respective ones of the N intermediate routing nodes to the at least one destination node in a disjoint manner such that no two paths between the source node and the at least one destination node have any nodes or links in common to prevent a single link failure or a single node failure from affecting the complete traffic flow between any source-destination node pair,
wherein a first routing step of the distributing step comprises finding a first set of disjoint paths from the source node to the respective ones of the N intermediate routing nodes, and
wherein a subsequent routing step of the distributing step comprises finding a second set of disjoint paths from each of the respective ones of the N intermediate routing nodes to the at least one destination node.
10. The apparatus of claim 9 , wherein the distributing operation further comprises determining one or more disjoint paths from the source node to at least a portion of the N intermediate routing nodes.
11. The apparatus of claim 9 , wherein the distributing operation further comprises determining one or more disjoint paths from at least a portion of the N intermediate routing nodes to the at least one destination node.
12. Apparatus for use in processing a traffic flow in a communication network comprising a plurality of nodes, the traffic flow being deliverable from a source node to at least one destination node via one or more intermediate routing nodes, the apparatus comprising:
a memory; and
a processor coupled to a memory, the processor being operative to split the traffic flow into a plurality of N parts, and to distribute the N parts to respective ones of a plurality of N intermediate routing nodes such that the parts are routed from the source node to the at least one destination node in a disjoint manner such that no two paths between the source node and the at least one destination node have any nodes or links in common to prevent a single link failure or a single node failure from affecting the complete traffic flow between any source-destination node pair;
wherein the distributing operation further comprises distributing, at most, two parts of the traffic flow on any given link of the network.
13. The apparatus of claim 9 , wherein distributing operation further comprises distributing the N parts of the traffic flow to the respective ones of the N intermediate routing nodes over pre-provisioned circuits each configured to support a corresponding one of the N parts.
14. The apparatus of claim 9 , wherein each of the N parts follows one of N different paths from the source node to the destination node via a corresponding one of the N intermediate routing nodes.
15. Apparatus for use in processing a traffic flow in a communication network comprising a plurality of nodes, the traffic flow being deliverable from a source node to at least one destination node via one or more intermediate routing nodes, the apparatus comprising:
a memory; and
a processor coupled to a memory, the processor being operative to: introduce two parity packets for every k−2 packets of traffic in the traffic flow at the source node, where k is the total number of packets; split the traffic flow with the parity packets into a plurality of k parts, and to distribute the k parts to respective ones of a plurality of intermediate routing nodes such that the k parts are routed from the source node to the at least one destination node in a disjoint manner where no two paths between the source node and the at least one destination node have any nodes or links in common to prevent a single link failure or a single node failure from affecting the complete traffic flow between any source-destination node pair.
16. An article of manufacture storing one or more programs for use in processing a traffic flow in a communication network comprising a plurality of nodes, the traffic flow being deliverable from a source node to at least one destination node via one or more intermediate routing nodes, the one or more programs when executed in a processor implementing a method comprising the steps of:
splitting the traffic flow at the source node into a plurality of N parts; and
distributing the N parts such that the N parts are first routed from the source node to respective ones of N intermediate routing nodes in a disjoint manner and then subsequently routed from the respective ones of the N intermediate routing nodes to the at least one destination node in a disjoint manner such that no two paths between the source node and the at least one destination node have any nodes or links in common to prevent a single link failure or a single node failure from affecting the complete traffic flow between any source-destination node pair,
wherein a first routing step of the distributing step comprises finding a first set of disjoint paths from the source node to the respective ones of the N intermediate routing nodes, and
wherein a subsequent routing step of the distributing step comprises finding a second set of disjoint paths from each of the respective ones of the N intermediate routing nodes to the at least one destination node.
17. The article of claim 16 , wherein the distributing step further comprises the step of determining one or more disjoint paths from the source node to at least a portion of the N intermediate routing nodes.
18. The article of claim 16 , wherein the distributing step further comprises the step of determining one or more disjoint paths from at least a portion of the N intermediate routing nodes to the at least one destination node.
19. An article of manufacture storing one or more programs for use in processing a traffic flow in a communication network comprising a plurality of nodes, the traffic flow being deliverable from a source node to at least one destination node via one or more intermediate routing nodes, the one or more programs when executed in a processor implementing a method comprising the steps of:
splitting the traffic flow at the source node into a plurality of N parts; and
distributing the N parts to respective ones of N intermediate routing nodes such that the N parts are routed from the source node to the at least one destination node in a disjoint manner such that no two paths between the source node and the at least one destination node have any nodes or links in common to prevent a single link failure or a single node failure from affecting the complete traffic flow between any source-destination node pair;
wherein the distributing step further comprises the step of distributing, at most, two parts of the traffic flow on any given link of the network.
20. An article of manufacture storing one or more programs for use in processing a traffic flow in a communication network comprising a plurality of nodes, the traffic flow being deliverable from a source node to at least one destination node via one or more intermediate nodes, the one or more programs when executed in a processor implementing a method comprising the steps of:
introducing two parity packets for every k−2 packets of traffic at the source node, where k is the total number of packets;
splitting the traffic flow with the parity packets at the source node into a plurality of k parts; and
distributing the k parts to respective ones of the intermediate nodes such that the k parts are routed from the source node to the at least one destination node in a disjoint manner where no two paths between the source node and the at least one destination node have any nodes or links in common to prevent a single link failure or a single node failure from affecting the complete traffic flow between any source-destination node pair.