COMBINATORIAL DESIGN METHOD AND APPARATUS FOR MULTI-RING NETWORKS WITH COMBINED ROUTING AND FLOW CONTROL

Patent №

US 6,563,833

Granted

2003-05-13

Filed 1999

Owner

LUCENT TECHNOLOGIES INC.

Lab

AI components

2

ml · hardware

Assignment

Recorded

Dataset

AIPD

2023_r1 edition

Application

09225991

A method and apparatus are disclosed for designing multi-ring networks based on generalized quadrangles of combinatorial design theory, together with a scaling algorithm for realizing networks of arbitrary size. Multi-ring networks based on generalized quadrangles ensures that the path between any two nodes is either confined to a single ring, or traverses exactly two rings (passing through a single bridge node), referred to herein as the “one-bridge property.” The one-bridge property allows the construction of networks with small rings and low degree, while also permitting efficient congestion-free flow control protocols. Routing, access and flow control protocols are provided for the multi-ring network that capitalize on the combinatorial properties of the disclosed multi-ring networks. A “self-routing” routing protocol is disclosed wherein a source node does not need to know how to route a packet to a destination node that is not a neighbor. The access control mechanism disclosed herein utilizes a quota counter, k, and allows each node to transmit up to the predefined quota of k packets on each ring, Ri, that the node belongs to, during any cycle. A set of flow control buffers, consisting of a local queue, LQi, a remote queue, RQij(x), and a buffer queue, BQij, and a flow control quota matrix, Qi, are utilized to implement a flow control mechanism. The routing, flow and access control protocols, together with the disclosed network architecture, ensure that (i) no loss due to congestion occurs inside a network, under arbitrary traffic patterns; (ii) all the packets reach their destinations within a bounded time; and (iii) the bandwidth is allocated fairly and no host is starved.

Machine learningAI hardwareH04L 47/29H04L 12/427H04L 45/04H04L 49/90

AI classification

Machine learning0.99
AI hardware0.77
Knowledge representation0.10
Evolutionary computation0.00
Vision0.00
Speech0.00
Natural language0.00
Planning0.00

Ownership

LUCENT TECHNOLOGIES INC.

assignment · 96910950

Assignors

WOOL, AVISHAI, YENER, BULENT

On an employer assignment, the assignors are typically the inventors.

© 2026 NYSGPT2525 LLC