Patent No. US10574562 (titled "Routing methods, systems, and computer program products") on Jun 26, 2019. The application was issued on Feb 25, 2020.
’562 is related to the field of network routing and packet forwarding, specifically within architectures that utilize segment routing to direct traffic through a network domain. Traditional routing often relies on complex state maintenance at every hop or global label distribution protocols that can struggle with scalability as networks grow. The background context involves the evolution of the Internet Protocol (IP) and the need for more efficient, path-based addressing mechanisms that can reduce latency and simplify the mapping between a node’s identity and the route taken to reach it.
The underlying idea behind ’562 is the dynamic derivation of segment identifiers through a mathematical combination of a stable, globally unique index and a locally significant base value. Rather than requiring every node to agree on a single global label for a destination, the invention allows each node to calculate a context-specific segment ID by adding a node-specific offset (the base value) to a universal index that represents the target nodal segment. This decoupling allows the network to maintain global reachability while respecting the local label space constraints of individual hardware components.
The claims of ’562 focus on a node configured to receive a globally unique index value and a local base value to compute both its own segment identifier and the identifier for the next hop in a path. The independent claims specifically cover the mechanism of calculating a next hop segment ID by applying the same global index to a second base value that has been reserved by the adjacent node. This ensures that as a packet traverses the network, the label used to identify the destination is automatically translated to match the expectations of the subsequent router in the sequence.
In practice, this system functions as a programmable forwarding plane where the path is encoded as a series of segments. When a packet arrives, the current node uses its internal segment routing global block (the base value) and the unique index to identify the traffic's destination within its own forwarding table. To move the packet forward, the node performs a label swap or imposition using a calculated value derived from the neighbor's base value, effectively performing a mathematical translation of the routing instruction that remains consistent across the entire domain.
This approach differs from prior solutions by eliminating the need for a centralized label distribution authority to synchronize every specific hop-by-hop label. By using a globally unique index as the constant variable in the routing equation, the invention allows for decentralized path computation. It cuts through the overhead of traditional MPLS or standard IP routing by enabling nodes to independently determine the correct forwarding labels for any destination, provided they know the base offsets used by their immediate neighbors.
In the early 2010s when ’562 was filed, network routing was typically implemented using a rigid distinction between names, which identified a resource, and addresses, which identified a topological location. At a time when systems commonly relied on global addressing schemes like IPv4 and IPv6, routing protocols were generally constrained by the need for large, centralized routing tables that mapped destination addresses to next-hop interfaces. Hardware and software constraints made the dynamic association of path-specific information directly within a protocol address non-trivial, as standard architectures were designed to process fixed-length, hierarchical address headers rather than variable or path-based identifiers.
The disclosed invention represents a technical advancement through an architectural shift that integrates path information directly into the protocol address structure. By utilizing a globally unique index value in combination with a base value to calculate segment identifiers, the system enables a node to dynamically determine both its own segment ID and the next-hop segment ID for a packet. This integration overcomes the technical constraints of traditional destination-based routing by allowing for segment-based path identification within a network domain. The resulting capability enables more efficient packet forwarding and path-specific routing without requiring the exhaustive state overhead typically associated with complex source-routing or large-scale global routing tables.
This patent contains 63 claims, with independent claims 1, 9, 18, and 63 focusing on a system, method, and apparatus for network routing that utilizes a globally unique index value and a base value to calculate segment identifiers and next-hop segment identifiers for packet forwarding. The independent claims specifically address the calculation of these identifiers based on segment routing values reserved by nodes within a network domain to identify nodal segments. The dependent claims further define these operations by specifying the use of locally-significant identifiers, packet header modifications, advertising index values and ranges, data structure storage, and the implementation of virtual identifier spaces to optimize network path determination and reduce control signaling.
Definitions of key terms used in the patent claims.
US Latest litigation cases involving this patent.

The dossier documents provide a comprehensive record of the patent's prosecution history - including filings, correspondence, and decisions made by patent offices - and are crucial for understanding the patent's legal journey and any challenges it may have faced during examination.
Get instant alerts for new documents