Metadata-Driven Serverless Function Placement Across Federated Platforms

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Serverless workloads in existing platforms are not federated and cannot be dynamically and intelligently placed, leading to inefficient execution.

Innovation Solution

Utilizing metadata to evaluate and dynamically place serverless workloads across edge-core-cloud deployments, considering factors like invocation source, data location, resource requirements, and execution preferences.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If serverless workloads are executed on existing serverless platforms, then function execution is provided, but workloads cannot be dynamically and intelligently placed across federated environments

Engineering Contradiction:

Improveworkload placement flexibilityVSAvoidplatform federation complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

An orchestration engine is introduced as an intermediary component that receives function requests, evaluates multiple serverless platforms using metadata, and intelligently selects the optimal platform for execution. This mediator handles the complexity of platform federation, allowing individual platforms to remain independent while enabling cross-platform workload placement through centralized coordination.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system implements feedback mechanisms where the orchestration engine continuously evaluates metadata from multiple serverless platforms, learns from execution outcomes, and dynamically adjusts platform selection decisions. This feedback loop enables intelligent adaptation to changing platform conditions and workload requirements, improving placement decisions over time.

Inventive Principle:

Principle #23Feedback

2Productivity

If metadata-based evaluation is implemented for function placement, then intelligent workload placement is achieved, but system complexity increases

Engineering Contradiction:

Improveworkload execution efficiencyVSAvoidmetadata evaluation complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The metadata evaluation process is segmented into distinct components: container metadata evaluation, node metadata evaluation, and data metadata evaluation. Each metadata type is processed independently by specialized evaluation logic, allowing the system to manage complexity through modular organization while comprehensively assessing platform suitability for workload placement.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Metadata is collected and evaluated in advance before workload execution decisions are made. The orchestration engine pre-assesses platform capabilities, resource availability, and data location information, so that when a function request arrives, the optimal platform can be quickly selected without complex real-time computation.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS12474970B2Intelligent function execution leveraging metadata

Publication Date: 2025.11.18 DELL PROD LP

AI summary

Executing functions in serverless platforms leveraging metadata is disclosed. Embodiments of the invention related to using metadata when placing and/or executing functions. The metadata includes request metadata, container or function metadata, node metadata, and/or data metadata. The metadata allows a function to be placed and executed in a manner that accounts for user intent, container requirements, node resources, and data location. Functions can be placed across different serverless platforms. Once the metadata is evaluated and a location is selected, a function is placed and executed at the selected location.