Disaggregated and distributed composable infrastructure

Patent metadata

Field	Value
Publication number	US-11425224-B2
Application number	US-202117184884-A
Country	US
Kind code	B2
Filing date	Feb 25, 2021
Priority date	Feb 25, 2020
Publication date	Aug 23, 2022
Grant date	Aug 23, 2022

Abstract

Novel tools and techniques are provided for implementing intent-based disaggregated and distributed composable infrastructure. In some embodiments, a computing system might receive, over a network, a request for network services from a customer, the request comprising desired characteristics and performance parameters, without specific information regarding any of hardware, hardware type, location, or network for providing the requested services. The computing system might identify network resources based at least in part on the desired characteristics and performance parameters, might establish transport links between the identified two or more network resources (which may be disaggregated and distributed), might configure (in some cases, based on derived distributable synchronization state(s)) at least one of the identified network resources to simulate zero (or near-zero) latency and/or to simulate zero (or near-zero) distance between the identified network resources, and might allocate the identified two or more network resources for providing the requested network services.

First claim

What is claimed is:

1. A method, comprising:
receiving, with a computing system over a network, a request for network services from a customer, the request for network services comprising desired characteristics and performance parameters for the requested network services, without information regarding any of specific hardware, specific hardware type, specific location, or specific network for providing the requested network services;
identifying, with the computing system, two or more network resources from two or more first networks capable of providing the requested network services, based at least in part on the desired characteristics and performance parameters for the requested network services;
establishing, with the computing system, one or more transport links between the identified two or more network resources, the identified two or more network resources being disaggregated and distributed network resources;
deriving, with the computing system, distributable synchronization state across at least one of the identified two or more network resources, a backplane of one or more of the two or more first networks, or the one or more transport links;
configuring, with the computing system, at least one network resource of the identified two or more network resources to perform at least one of simulating zero latency or near-zero latency between the identified two or more network resources or simulating zero distance or near-zero distance between the identified two or more network resources, based at least in part on the desired characteristics and performance parameters for the requested network services and based at least in part on the derived distributable synchronization state; and
allocating, with the computing system, the identified two or more network resources for providing the requested network services,
wherein deriving, with the computing system, distributable synchronization state across at least one of the identified two or more network resources, a backplane of one or more of the two or more first networks, or the one or more transport links comprises performing one of:
comparing, with the computing system, system clocks each associated with each of the identified two or more network resources, and deriving, with the computing system, the distributable synchronization state based on any differences in the comparison of the system clocks; or
comparing, with the computing system, two or more Qbit multi-states of two or more quantum timing systems associated with at least two of the identified two or more network resources, and deriving, with the computing system, the distributable synchronization state based on any differences in the comparison of the two or more Qbit multi-states of each quantum timing system.

2. The method of claim 1 , wherein the computing system comprises one of a path computation engine, a data flow manager, a server computer over a network, a cloud-based computing system over a network, or a distributed computing system.

3. The method of claim 1 , wherein the one or more transport links comprise at least one of one or more optical transport links, one or more network transport links, or one or more wired transport links.

4. The method of claim 1 , wherein simulating zero latency or near-zero latency between the identified two or more network resources comprises using a re-timer to simulate zero latency or near-zero latency between the identified two or more network resources, based at least in part on the derived distributable synchronization state.

5. The method of claim 1 , wherein simulating zero distance or near-zero distance between the identified two or more network resources comprises using a re-driver or a repeater to simulate zero distance or near-zero distance between the identified two or more network resources, based at least in part on the derived distributable synchronization state.

6. The method of claim 1 , wherein simulating zero latency or near-zero latency between the identified two or more network resources or simulating zero distance or near-zero distance between the identified two or more network resources comprises utilizing a buffer with flexible buffer capacity to simulate zero latency or near-zero latency between the identified two or more network resources or to simulate zero distance or near-zero distance between the identified two or more network resources, based at least in part on the derived distributable synchronization state.

7. The method of claim 1 , wherein establishing the one or more transport links between the disaggregated and distributed identified two or more network resources comprises utilizing light steered transport to establish the one or more transport links between the disaggregated and distributed identified two or more network resources.

8. The method of claim 1 , further comprising:
mapping, with computing system, a plurality of network resources within the two or more first networks;
wherein identifying the two or more network resources comprises identifying, with the computing system, the two or more network resources from the two or more first networks capable of providing the requested network services, based at least in part on the desired characteristics and performance parameters for the requested network services and based at least in part on the mapping of the plurality of network resources.

9. The method of claim 8 , wherein at least one of identifying the two or more network resources, mapping the plurality of network resources, or configuring the at least one network resource of the identified two or more network resources is performed using at least one of one or more artificial intelligence (“AI”) systems, one or more machine learning systems, or one or more software defined network (“SDN”) systems.

10. The method of claim 1 , wherein the identified two or more network resources comprise peripheral component interconnect (“PCI”) -based network cards each comprising one or more network interface cards (“NICs”), one or more smart NICs, one or more graphics processing units (“GPUs”), or one or more storage devices.

11. The method of claim 1 , wherein the identified two or more network resources comprise two or more generic or single-purpose network devices in place of specialized or all-purpose network devices.

12. The method of claim 1 , wherein the desired characteristics comprise at least one of requirement for network equipment to be geophysically proximate to the customer, requirement for network equipment to be located within a geophysical location, requirement to avoid routing network traffic through a geophysical location, requirement to route network traffic through a geophysical location, requirement to exclude a first type of network resources from fulfillment of the requested network services, requirement to include a second type of network resources for fulfillment of the requested network services, requirement to fulfill the requested network services based on a single goal indicated by the customer, or requirement to fulfill the requested network services based on multi-goals indicated by the customer.

13. The method of claim 1 , wherein the desired performance parameters comprise at least one of a maximum latency, a maximum jitter, a maximum packet loss, a maximum number of hops, performance parameters defined in a service level agreement (“SLA”) associated with the customer or performance parameters defined in terms of natural resource usage, quality of service (“QoS”) measurement data, platform resource data and metrics, service usage data, topology and reference data, historical network data, network usage trend data, or one or more of inform

Assignees

• Level 3 Communications Llc

Inventors

• Mcbride Kevin M
• Sutherland James E
• Moss Frank
• Smith Brent
• Stallings Charles
• Mollard Mitch
• O'Brien Jr William

Classifications

• H04L67/141

  Setup of application sessions (admission control or resource allocation in data switching networks H04L47/70) · CPC title

• H04L67/10 Primary

  in which an application is distributed across nodes in the network (software deployment G06F8/60; multiprogramming arrangements G06F9/46) · CPC title

• H04L67/60 Primary

  Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources (admission control or resource allocation H04L47/70) · CPC title

• H04L67/51

  Discovery or management thereof, e.g. service location protocol [SLP] or web services · CPC title