Resource-Negotiated Workload Orchestration for Deterministic Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for orchestrating control tasks in industrial plants lack flexibility in virtualized computing environments, particularly in ensuring deterministic computation and communication, which are critical for maintaining precise control and synchronicity.
Innovation Solution
A method for orchestrating the deterministic execution of workloads in computing platforms by determining resource consumption and translating it into performance requirements, allowing negotiation with management entities for optimal compute instance configurations that meet timing, synchronicity, and availability needs, enabling flexible deployment of control functions in virtualized environments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If control tasks are executed on dedicated hardware to guarantee deterministic computation and communication, then reliability and timing precision are improved, but adaptability and flexibility deteriorate
Solution Approach 1:
The system segments control functions into independent, deployable units that can be allocated to different compute instances. This segmentation enables flexible deployment while maintaining deterministic execution guarantees for each control task through isolated resource allocation.
Solution Approach 2:
The patent introduces an intermediary layer (orchestration system/resource manager) that sits between the control functions and the underlying hardware resources. This intermediary manages resource allocation and scheduling, providing deterministic execution guarantees to control tasks while allowing flexible deployment on virtualized infrastructure.
2Adaptability or versatility
If control functions are virtualized to increase flexibility and scalability, then adaptability is improved, but device complexity and difficulty in ensuring deterministic execution increase
Solution Approach 1:
The patent creates a universal orchestration framework that can manage multiple types of compute instances (virtual machines, containers, bare-metal) through a common interface and resource allocation mechanism. This universal approach simplifies the complexity of managing diverse virtualized environments while maintaining deterministic execution capabilities.
Solution Approach 2:
The system dynamically adjusts resource allocation parameters (CPU time slices, memory allocation, I/O bandwidth) to ensure deterministic execution in virtualized environments. By carefully controlling these parameters, the patent maintains timing guarantees while operating in flexible virtualized infrastructures.
3Productivity
If resource allocation is optimized for performance requirements, then productivity is improved, but device complexity increases due to negotiation and monitoring mechanisms
Solution Approach 1:
The patent implements preliminary action by pre-negotiating and reserving resource allocations before control tasks execute. Resource profiles are established in advance, specifying CPU time, memory, and I/O bandwidth guarantees. This preliminary resource reservation simplifies runtime management while ensuring deterministic performance.
Solution Approach 2:
The system incorporates feedback mechanisms that monitor resource consumption and execution performance, using this information to dynamically adjust resource allocation. This feedback loop optimizes productivity while maintaining deterministic guarantees, and the automated nature of the feedback reduces manual complexity.
Data Source
AI summary
A computer-implemented method for orchestrating the deterministic execution of a given workload on at least one computing platform, comprising determining a consumption of at least one computing resource and a consumption of at least one communications resource that result from executing a given workload or a part thereof; determining, from these consumptions and a given set of requirements that relate to the timing, synchronicity and/or availability of executing the given workload or part thereof, at least one performance requirement with respect to execution of the workload or part thereof; and negotiating, with a management entity of the computing platform, execution of the workload or part thereof on the computing platform according to the at least one performance requirement.