Granted Patent B2

US 12,519,733 · App. 17/561,886 · Granted Jan 6, 2026

Inventors: Jing Zhu (Portland, OR); Menglei Zhang (Portland, OR); Wey-Yi Guy (Beaverton, OR); Valerie Parker (Portland, OR)

Assignee: Intel Corporation

H04L47/2483H04L47/2441H04L47/2475H04L47/283H04L47/34

Abstract

Claims (52)

1 . A multi-access (MX) compute node to operate in an MX computing environment, the MX compute node comprising:

memory circuitry; and

processor circuitry connected to the memory circuitry, the processor circuitry to execute instructions of an MX traffic shaper, which, when executed are to cause the processor circuitry to:

classify incoming packets into a traffic class (TC) of a plurality of TCs;

select, for each packet of the incoming packets, one or more access networks of a plurality of access networks to deliver the incoming packets to a destination device according to the TC of each packet;

store each packet in a corresponding queue of a plurality of queues according to the TC of each packet such that each packet having a same TC is stored in a same queue of the plurality of queues, wherein the plurality of queues includes a set of queues for each access network of the plurality of access networks, and each set of queues includes a queue corresponding to each TC in the plurality of TCs;

schedule packets for delivery to the selected one or more access networks based on a TC of each queue of the plurality of queues and based on the access network associated with each queue; and

cause delivery of the queued packets to the one or more access networks based on the scheduling.

2 . The MX compute node of claim 1 , wherein each TC of the plurality of TCs corresponds to a quality of service (QOS) class, priority level, or a set of application requirements.

3 . The MX compute node of claim 1 , wherein, to select the one or more access networks, execution of the instructions is to cause the processor circuitry to:

select one access network of the plurality of access networks for each packet classified as belonging to a first TC of the plurality of TCs; and

select multiple access networks of the plurality of access networks for each packet classified as belonging to a second TC of the plurality of TCs.

4 . The MX compute node of claim 1 , wherein execution of the instructions is to cause the processor circuitry to:

select one access network of the plurality of access networks for a set of the incoming packets classified as belonging to a first TC of the plurality of TCs; and

select multiple access networks of the plurality of access networks for a set of the incoming packets classified as belonging to a second TC of the plurality of TCs such that different access networks of the plurality of access networks are selected for different subsets of the set of incoming packets classified as belonging to the second TC.

5 . The MX compute node of claim 1 , wherein the plurality of queues includes a set of queues for each access network of the plurality of access networks, and each set of queues includes a queue corresponding to a TC in the plurality of TCs.

6 . The MX compute node of claim 1 , wherein execution of the instructions is to cause the processor circuitry to:

determine a number of TCs in the plurality of TCs based on a received traffic shaping configuration, wherein the traffic shaping configuration includes, for each TC in the plurality of TCs, a TC identifier (ID), a set of TC parameters, and a set of TC-specific tunneling parameters.

7 . The MX compute node of claim 6 , wherein the set of TC parameters includes a source network address, a destination network addresses, a protocol type, source port number, and a destination port number.

8 . The MX compute node of claim 6 , wherein:

the traffic shaping configuration includes, for each TC in the plurality of TCs, a number of transmission schedules for each TC, wherein each transmission schedule in the number of transmission schedules indicates a period during which transmission are to take place, and for each transmission schedule, the traffic shaping configuration includes a start time, a transmission duration, and a transmission period; or

the traffic shaping configuration includes, for each TC in the plurality of TCs, a number of no-transmission schedules for each TC, wherein each no-transmission schedule in the number of no-transmission schedules indicates a period during which no transmissions are to take place, and for each no-transmission schedule, the traffic shaping configuration includes a start time, a no-transmission duration, and a no-transmission period.

9 . The MX compute node of claim 6 , wherein the traffic shaping configuration includes, for each access network in the plurality of access networks, a transmission rate for scheduling transmissions and a queueing mode for scheduling transmissions.

10 . The MX compute node of claim 9 , wherein the traffic shaping configuration includes, for each TC of the plurality of TCs, a TC ID, and one of:

a priority assigned to a queue of the plurality of queues corresponding to the TC ID when the queueing mode is a strict-priority mode; and

a weighted round robin (WRR) weight assigned to a queue of the plurality of queues corresponding to the TC ID when the queueing mode is a WRR mode.

11 . The MX compute node of claim 1 , wherein the MX traffic shaper is part of a convergence layer in a Multi-Access Management Services (MAMS) control plane protocol stack.

12 . The MX compute node of claim 1 , wherein execution of the instructions is to cause the MX compute node to:

classify incoming packets into a traffic class (TC) of a plurality of TCs;

select, for each packet of the incoming packets, one or more access networks of a plurality of access networks to deliver the incoming packets to a destination device according to the TC of each packet;

store each packet in a corresponding queue of a plurality of queues according to the TC of each packet such that each packet having a same TC is stored in a same queue of the plurality of queues;

schedule packets for delivery to the selected one or more access networks based on a TC of each queue of the plurality of queues; and

cause delivery of the queued packets to the one or more access networks based on the scheduling.

13 . The MX compute node of claim 1 , wherein the MX compute node is a client device.

14 . One or more non-transitory computer readable media (NTCRM) comprising instructions, wherein execution of the instructions by one or more processors of a multi-access (MX) compute node is to cause the MX compute node to:

determine to operate an MX traffic shaping function when a peer MX compute node indicates support for an MX traffic shaping capability in a capability message; and

when the peer MX compute node indicates support for the MX traffic shaping capability,

classify incoming packets into a traffic class (TC) of a plurality of TCs;

selecting, for each packet of the incoming packets, one or more access networks of a plurality of access networks to deliver the incoming packets to a destination device according to the TC of each packet;

storing each packet in a corresponding queue of a plurality of queues according to the TC of each packet such that each packet having a same TC is stored in a same queue of the plurality of queues, wherein the plurality of queues includes a set of queues for each access network of the plurality of access networks, and each set of queues includes a queue corresponding to each TC in the plurality of TCs;

scheduling packets for delivery to the selected one or more access networks based on a TC of each queue of the plurality of queues and based on the access network associated with each queue; and

cause delivery of the queued packets to the one or more access networks based on the scheduling.

15 . The one or more NTCRM of claim 14 , wherein execution of the instructions is to cause the MX compute node to:

determine whether the peer MX compute node supports a TC-specific tunneling capability based on the information included in the capability message, wherein, when the peer MX compute node supports the TC-specific tunneling capability, the MX traffic shaping function includes, for each TC in the plurality of TCs, a set of TC-specific tunneling parameters, and the set of TC-specific tunneling parameters includes a transport port number and a QoS class or treatment;

establish a dedicated transport tunnel with the peer MX compute node for communicating packets having a TC indicated by the set of TC-specific tunneling parameters; and

map a quality of service (QOS) treatment to a transport flow to be communicated through the dedicated transport tunnel.

16 . The one or more NTCRM of claim 15 , wherein execution of the instructions is to cause the MX compute node to:

determine whether the peer MX compute node supports a time synchronization (sync) capability based on the information included in a capability message;

receive a time sync configuration message when the peer MX compute node supports the time sync capability;

perform one or more time sync operations based on information included in the time sync configuration message; and

send a time sync configuration response message to the peer MX node including an indicator indicating success or failure of the one or more time sync operations.

17 . The one or more NTCRM of claim 16 , wherein the time sync configuration includes a time sync transport type, a time sync protocol, a time sync network address, and a time sync port number.

Assignments (1)

Continuity (1)

References Cited (165)