Data Plane API for Autonomous Resource Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In distributed computing systems, existing technologies require commands to traverse the entire cloud service stack for configuration and management actions, leading to inefficiencies and delays, as well as limited autonomy for nodes in making local decisions.
Innovation Solution
Implementing a control plane application programming interface (API) that allows granular execution at the virtual machine level, enabling local API calls to manage resources without requiring commands from the control plane, allowing nodes to make decisions such as scaling, restarting, and attaching/detaching resources within specified limits.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If commands traverse the entire cloud service stack for configuration and management actions, then centralized control is maintained, but system responsiveness and operational efficiency deteriorate
Solution Approach 1:
The patent segments the cloud service stack into multiple hierarchical levels (control plane at top, regional planes in middle, data plane at bottom). Each level handles specific operations locally without requiring commands to traverse the entire stack. This segmentation enables local decision-making at data plane nodes while maintaining centralized policy enforcement at the control plane, thereby improving responsiveness without oversimplifying the overall control architecture.
Solution Approach 2:
The patent introduces a hierarchical dimension to the control architecture, organizing control planes at multiple levels (control plane, regional planes, data plane). This dimensional organization allows operations to be resolved at the appropriate hierarchical level, reducing the need for commands to traverse the entire stack while maintaining centralized coordination through the hierarchical structure.
2Productivity
If nodes have autonomous decision-making capability, then operational efficiency improves, but system-wide coordination and control become more difficult
Solution Approach 1:
The patent implements feedback mechanisms where data plane nodes report status and decisions to regional planes, which in turn report to the control plane. This hierarchical feedback system enables autonomous nodes to make local decisions while maintaining system-wide coordination through structured information flow. The feedback loops ensure that autonomous operations align with overall system policies and goals.
Solution Approach 2:
The patent introduces regional planes as intermediary layers between the control plane and data plane nodes. These intermediaries handle local coordination and decision-making, reducing the burden on the control plane while ensuring system-wide coordination. The regional planes act as mediators that translate control plane policies into local actions and aggregate local status information for upward reporting.
3Productivity
If control commands traverse the entire stack, then centralized authorization is ensured, but deployment time and resource provisioning speed increase
Solution Approach 1:
The patent implements preliminary authorization where the control plane pre-approves resource allocation limits and policies before deployment. Data plane nodes operate within these pre-established boundaries, making local decisions without requiring real-time control plane approval for each action. This preliminary action ensures authorization control is maintained while enabling rapid resource provisioning.
Solution Approach 2:
The patent applies partial authorization by granting data plane nodes limited autonomous decision-making power within predefined boundaries. Nodes can perform certain operations locally without full control plane involvement, while still operating under the umbrella of centralized authorization policies. This partial delegation of authority speeds up resource provisioning while maintaining adequate authorization control.
Data Source
AI summary
Embodiments are directed to a distributed computing system comprising a plurality of compute nodes for providing resources to users and a hierarchy of two or more layers of controllers coupling the compute nodes to a user interface via a control plane, wherein at least one compute node receives a local application program interface (API) call from an application running on the at least one compute node, the local API call causing the at least one compute node to configure a local resource without requiring commands from the control plane.