LwM2M Hardware Discovery for Constrained IoT Orchestration

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current orchestration technologies for distributed IoT computing environments are inefficient due to excessive memory and computational requirements, inability to manage hardware resources of constrained devices, and lack of mechanisms for orchestrating Unikernel-based services, limiting their use in IoT ecosystems.

Innovation Solution

Modifying the LwM2M protocol to determine hardware characteristics of IoT devices and using CoAP as an application protocol with LwM2M as an orchestration management tool, introducing HW-INFO and UNIKERNEL-DATA objects to expose hardware information and manage Unikernel applications, enabling adaptive service deployment and execution.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If orchestration technologies are deployed in distributed IoT computing environments, then service management capability is improved, but memory and computational requirements increase excessively

Engineering Contradiction:
Improveservice management capabilityVSAvoidmemory and computational requirements
Core Design Contradiction:
Extent of automationVSQuantity of substance

Solution Approach 1:

The patent extracts the essential hardware characteristics (CPU frequency, memory size, storage capacity) from complex device specifications and represents them through simplified LwM2M data models. This extraction allows orchestration to work with minimal essential data rather than complete device states, reducing computational overhead while maintaining service management capability.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

Instead of having the orchestrator query detailed device information continuously, the patent inverts the approach by having devices autonomously publish their hardware characteristics and service capabilities through LwM2M. This inversion reduces the computational burden on the orchestrator and enables constraint ed devices to participate in distributed computing with minimal resource consumption.

Inventive Principle:
Principle #13The other way round (Inversion)

2Adaptability or versatility

If traditional virtualization technologies are used in IoT environments, then service deployment flexibility is improved, but device resource overhead increases

Engineering Contradiction:
Improveservice deployment flexibilityVSAvoiddevice resource overhead
Core Design Contradiction:
Adaptability or versatilityVSQuantity of substance

Solution Approach 1:

The patent employs lightweight unikernel-based virtual machine images that are minimal in size and can be rapidly deployed and replaced. These disposable VM images contain only the essential service code and dependencies, eliminating the need for heavy traditional virtualization layers. Services can be quickly instantiated, migrated, or terminated based on orchestration decisions without significant resource overhead.

Inventive Principle:
Principle #27Cheap short-living objects (Disposable)

Solution Approach 2:

The patent changes the virtualization parameter from heavy full-featured OS images to lightweight unikernel images with minimal footprints. By modifying the virtualization approach to use compiled, statically-linked binary images rather than dynamic OS installations, the system achieves service deployment flexibility while reducing device resource overhead by orders of magnitude.

Inventive Principle:
Principle #35Parameter changes

3Loss of information

If LwM2M protocol is extended with HW-INFO and UNIKERNEL-DATA objects, then hardware resource visibility is improved, but protocol complexity increases

Engineering Contradiction:
Improvehardware resource visibilityVSAvoidprotocol complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The patent segments hardware information into distinct LwM2M data objects: HW-INFO for device characteristics (CPU, memory, storage) and UNIKERNEL-DATA for virtual machine service capabilities. This segmentation allows the protocol to expose comprehensive hardware resource visibility while maintaining structured, modular complexity that is manageable and extensible.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent creates universal LwM2M data objects that serve multiple functions: HW-INFO objects provide hardware characteristics for service placement decisions, while UNIKERNEL-DATA objects provide both service capability information and resource requirements. This multi-functionality reduces the need for separate protocol mechanisms, achieving comprehensive hardware visibility without proportionally increasing protocol complexity.

Inventive Principle:
Principle #6Universality (Multi-functionality)

4Power

If Unikernel-based services are deployed on constrained devices, then processing performance is improved, but lack of orchestration mechanisms limits adaptability

Engineering Contradiction:
Improveprocessing performanceVSAvoidorchestration capability
Core Design Contradiction:
PowerVSAdaptability or versatility

Solution Approach 1:

The patent introduces LwM2M as an intermediary layer between unikernel-based services and the orchestration system. This mediator enables constrained devices running unikernels to expose their hardware characteristics and service capabilities through standardized data models, allowing adaptive orchestration decisions without requiring complex native orchestration mechanisms on each device.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent implements preliminary action by having devices pre-publish their hardware characteristics (CPU frequency, memory, storage) and service capabilities through LwM2M before orchestration decisions are made. This advance information availability enables the orchestrator to make informed placement and migration decisions, achieving adaptability without requiring real-time complex interactions with constrained unikernel devices.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP4055849B1Distributed computation orchestration for internet-of-things devices using coap and LWM2m protocols
Publication Date: 2024.09.04 TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)
  • EP4055849B1 patent drawingFigure 1
  • EP4055849B1 patent drawingFigure 2
  • EP4055849B1 patent drawingFigure 3

AI summary

An IoT electronic device executes services distributed by an IoT service orchestration device. A Lightweight Machine-to-Machine (LwM2M) request message is received. TheLwM2M request message contains a LwM2M object identifying hardware resources of the IoT electronic device for which characteristics are requested. A LwM2M command is executed that accesses a LwM2M interface identified based on content of the LwM2M object to determine the characteristics of the hardware resources of the IoT electronic device which are identified by the LwM2M object. A response message contains information identifying the characteristics of the hardware resources of the IoT electronic device. The response message is communicated toward the IoT service orchestrator device. A service image is received for execution which is adapted by the IoT service orchestrator device, responsive to the information in the response message identifying the characteristics of the hardware resources of the IoT electronic device.