IP Library › Granted Patent US 12,464,051
Granted Patent B2
US 12,464,051 · App. 17/712,119 · Granted Nov 4, 2025

Queuing control for distributed compute network orchestration

Inventors: Hao Feng (Hillsboro, OR); Yi Zhang (San Jose, CA); Srikathyayani Srikanteswara (Portland, OR); Marcin Spoczynski (Lexilip, IE); Nageen Himayat (Fremont, CA); Alexander Bachmutsky (Sunnyvale, CA); Maruti Gupta Hyde (Portland, OR)
Assignee: Intel Corporation
H04L67/51H04L67/63
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,464,051
App. No.
17/712,119
Granted
Nov 4, 2025
Kind
B2
Abstract

In one embodiment, a node of a data centric network (DCN) may receive a first service request interest packet from another node of the DCN, the first service request interest packet indicating a set of functions to be performed on source data to implement a service. The node may determine that it can perform a particular function of the set of functions, and determine, based on a backlog information corresponding to the particular function, whether to commit to performing the particular function or to forward the service request interest packet to another node. The node may make the determination further based on service delivery information indicating, for each face of the node, a service delivery distance for implementing the set of functions.

Claims (63)

1 . An apparatus of node of a data centric network (DCN), the apparatus including an interconnect interface to connect the apparatus to one or more components of the node, and a processor to:

receive a first service request interest packet from another node of the DCN, the first service request interest packet indicating a set of functions to be performed on source data to implement a service;

determine that the node can perform a particular function of the set of functions;

determine, based on a local backlog corresponding to the particular function, and backlogs of one or more other nodes of the DCN corresponding to the particular function, and service delivery information indicating, for each face of the node, a service delivery distance for implementing the set of functions, whether to commit to performing the particular function or to forward the service request interest packet to another node; and

cause a second service request interest packet to be transmitted to another node of the DCN based on the determination.

2 . The apparatus of claim 1 , wherein the service delivery distance is based on a number of network hops in the DCN for implementing the set of functions indicated by the second service request interest packet.

3 . The apparatus of claim 2 , wherein the number of network hops includes hops between nodes of the DCN and processing hops in the DCN corresponding to the performance of a function in the set of functions.

4 . The apparatus of claim 1 , wherein the processor is to obtain the service delivery information by:

receiving a service discovery interest packet from another node of the DCN, the service discovery interest packet indicating the set of functions;

determining that the node can perform a particular function;

causing a new service discovery interest packet to be transmitted to one or more other nodes of the DCN, the new service discovery interest packet indicating the set of functions without the particular function; and

receiving, from one or more other nodes of the DCN, discovery data packets indicating the service delivery distance information for implementing the set of functions.

5 . The apparatus of claim 1 , wherein the processor is to determine whether to commit to performing the particular function or to forward the service request interest packet to another node using on a backpressure routing algorithm based on Lyapunov drift optimization.

6 . The apparatus of claim 1 , wherein the processor is, based on a determination to commit to performing the particular function, generate the second service request interest packet to indicate the set of functions without the particular function.

7 . The apparatus of claim 1 , wherein the processor is, based on a determination to forward the services request interest packet to another node, generate the second service request interest packet to indicate the set of functions with the particular function.

8 . The apparatus of claim 1 , wherein the processor is further to receive periodic messages from the other nodes of the DCN indicating their respective backlogs corresponding to the particular function.

9 . The apparatus of claim 1 , wherein the processor is further to:

obtain a service data packet from another node of the DCN;

perform the first function on the service data packet; and

cause output data of the first function to be transmitted to another node of the network in a service data packet.

10 . One or more computer-readable media comprising instructions that, when executed by one or more processors of a node of a data centric network (DCN), cause the one or more processors to:

receive a first service request interest packet from another node of the DCN, the first service request interest packet indicating a set of functions to be performed on source data to implement a service;

determine that the node can perform a particular function of the set of functions;

determine, based on a local backlog corresponding to the particular function, and backlogs of one or more other nodes of the DCN corresponding to the particular function, and service delivery information indicating, for each face of the node, a service delivery distance for implementing the set of functions, whether to commit to performing the particular function or to forward the service request interest packet to another node; and

cause a second service request interest packet to be transmitted to another node of the DCN based on the determination.

11 . The computer-readable media of claim 10 , wherein the service delivery distance is based on a number of network hops in the DCN for implementing the set of functions indicated by the second service request interest packet.

12 . The computer-readable media of claim 11 , wherein the number of network hops includes hops between nodes of the DCN and processing hops in the DCN corresponding to the performance of a function in the set of functions.

13 . The computer-readable media of claim 10 , wherein the instructions are to obtain the service delivery information by:

receiving a service discovery interest packet from another node of the DCN, the service discovery interest packet indicating the set of functions;

determining that the node can perform a particular function;

causing a new service discovery interest packet to be transmitted to one or more other nodes of the DCN, the new service discovery interest packet indicating the set of functions without the particular function; and

receiving, from one or more other nodes of the DCN, discovery data packets indicating the service delivery distance information for implementing the set of functions.

14 . The computer-readable media of claim 10 , wherein the instructions are to determine whether to commit to performing the particular function or to forward the service request interest packet to another node using on a backpressure routing algorithm based on Lyapunov drift optimization.

15 . The computer-readable media of claim 10 , wherein the instructions are, based on a determination to commit to performing the particular function, generate the second service request interest packet to indicate the set of functions without the particular function.

16 . The computer-readable media of claim 10 , wherein the instructions are, based on a determination to forward the services request interest packet to another node, generate the second service request interest packet to indicate the set of functions with the particular function.

17 . The computer-readable media of claim 10 , wherein the instructions are further to:

receive a service data packet from another node of the DCN;

perform the first function on the service data packet; and

cause output data of the first function to be transmitted to another node of the network in a service data packet.

18 . A method to be performed by a node of a data centric network (DCN), comprising:

receiving a first service request interest packet, the first service request interest packet indicating a sequence of functions, in reverse order, to be executed to implement at least a portion of a service on source data;

orchestrating execution of a first function of the sequence of functions, comprising:

determining that the node can perform the first function;

committing to execution of the first function based on local backlog information corresponding to the first function, and backlog information of one or more other nodes of the DCN corresponding to the first function, and service delivery information indicating, for each face of the node, a service delivery distance for implementing the first function; and

transmitting one or more second service request interest packets to other nodes of the DCN indicate the sequence of functions, in reverse order, without the first function.

19 . The method of claim 18 , further comprising:

receiving a third service request interest packet indicating a sequence of functions in reverse order;

forwarding the third service request interest packet to one or more other nodes of the DCN for execution of the first function based on the local backlog information corresponding to the first function and the backlog information of one or more other nodes of the DCN corresponding to the first function.

20 . The method of claim 19 , further comprising:

performing a service discovery procedure in the DCN, comprising:

transmitting a service discovery interest packet to one or more other nodes of the DCN, the service discovery interest packets indicating the sequence of functions, in reverse order, to be implemented to perform at least a portion of the service on source data; and

receiving, from the one or more other nodes of the DCN, a discovery data packet indicating service delivery distance information for implementing the sequence of functions;

wherein the forwarding of the third service request interest packet comprises:

selecting a node of the DCN based on the service delivery distance information; and

transmitting the third service request interest packet to the selected node.

21 . The method of claim 20 , further comprising:

computing local abstract backlog information corresponding to the first function for the node and computing abstract backlog information of one or more other nodes of the DCN corresponding to the first function based on the service delivery distance information;

wherein the node selection is based on the local abstract backlog information and the abstract backlog information of one or more other nodes of the DCN.

22 . The method of claim 20 , wherein the service delivery distance information indicates a shortest path segment in the DCN from the node to a data producer node of the DCN.

23 . The method of claim 18 , further comprising:

receiving a service data packet from another node of the DCN;

performing the first function on data of the service data packet; and

transmitting output data of the first function to another node of the network in a service data packet.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded May 2, 2022
From: FENG, HAO; ZHANG, YI; SPOCZYNSKI, MARCIN; SRIKANTESWARA, SRIKATHYAYANI; HIMAYAT, NAGEEN; BACHMUTSKY, ALEXANDER; HYDE, MARUTI GUPTA
To: INTEL CORPORATION
Reel/Frame 059763/0364 →
Continuity (1)
Related Publication 20220224762A1 · Jul 14, 2022
References Cited (15)
US 10623321B2 · Mehra · 2020 [cited by examiner]
US 11853303B1 · Oliner · 2023 [cited by examiner]
US 20210409335A1 · Zhu · 2021 [cited by examiner]
Distributed Lyapunov Drift-plus-penalty Routing for WiFi Mesh Networks with Adaptive Penalty Weight by Jose Nunez-Martinez and Josep Mangues-Bafalluy (Year: 2012). [cited by examiner]
Afanasyev, Alexander, et. al., “NDF Developer's Guide,” NDN, Technical Report NDN-0021; Aug. 2021; https://named-data.net/publications/techreports/; 76 pages. [cited by applicant]
Alawi, Mahmoud A., et al., “Simplified gateway selection scheme for multihop relay in vehicular ad hoc network,” International Journal of Communication Systems, vol. 27, No. 12, 2014; pp. 3855-3873, 20 pages. [cited by applicant]
Barcelo, Marc, et al., “IoT-cloud service optimization in next generation smart environments”, IEEE Journal on Selected Areas in Communications, vol. 34, No. 12, 2016; pp. 4077-4090; 14 pages. [cited by applicant]
Bari, Md Faizul, et al., “On orchestrating virtual network functions,” in Proc. IEEE CNSM, 2015, pp. 50-56, 7 pages. [cited by applicant]
Kamran, Khashayar, et al., “Deco: Joint computation, caching and forwarding in data-centric computing networks”, in Proc. ACM Mobihoc, 2019, pp. 111-120, 10 pages. [cited by applicant]
Krol, Michael, et al., “Compute First Networking: Distributed Computing Meets ICN,” in Proc. ACM ICN, 2019, pp. 67-77, 11 pages. [cited by applicant]
Król, Michal, et al., “NFaaS: Named Function as a Service”, in Proc. ACM Conference on Information-Centric Networking, 2017, pp. 134-144, 11 pages. [cited by applicant]
Krol, Michal, et al., “RICE: Remote Method Invocation in ICN”, in ACM ICN, 2018, pp. 1-11, 11 pages. [cited by applicant]
Setiawan, Fudhiyanto Pranata, et al., “An optimum multiple metrics gateway selection mechanism in MANET and infrastructure networks integration,” in Proc. IEEE WCNC, 2008, pp. 2229-2234, 7 pages. [cited by applicant]
Topcuoglu, Haluk, et al., “Performance-effective and low complexity task scheduling for heterogeneous computing”, IEEE Transactions on Parallel and Distributed Systems, vol. 13, No. 3, pp. 260-274, 2002, 15 pages. [cited by applicant]
Tschudin , Christian, et al., “Named functions and cached computations,” in Proc. IEEE CCNC, 2014, pp. 851-857, 6 pages. [cited by applicant]