Dynamic Resource Rebalancing for Fog Computing Video Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Efficiently orchestrating large-scale computing applications across diverse and evolving distributed computing resources in fog computing systems is challenging due to complexity, heterogeneity, and varying workload demands, leading to issues like video frame dropping in video streaming applications.
Innovation Solution
Implementing a dynamic resource rebalancing mechanism in fog computing systems using low-latency persistent storage and a scalable peer selection algorithm to redirect video processing tasks among edge nodes, ensuring continuous processing without frame loss.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If resources are statically allocated to edge nodes for video processing, then device complexity is reduced, but video frame dropping occurs when workload exceeds allocated resources
Solution Approach 1:
The patent implements dynamic resource allocation where the orchestrator continuously monitors workload demands and rebalances video processing tasks across edge nodes in real-time. This allows the system to adapt to varying workload conditions and prevent frame dropping by redistributing tasks from overloaded nodes to underutilized nodes, transforming the static resource allocation into a dynamic system that responds to changing conditions.
Solution Approach 2:
The system employs feedback mechanisms where the orchestrator receives status information from edge nodes about their current workload and resource utilization. Based on this feedback, the orchestrator makes informed decisions about task redistribution to optimize resource utilization and maintain reliable video frame delivery, creating a closed-loop control system.
2Productivity
If video processing tasks are concentrated on fewer edge nodes, then processing efficiency is improved, but system reliability decreases due to single points of failure
Solution Approach 1:
The patent segments video processing tasks into smaller units that can be independently distributed across multiple edge nodes. The orchestrator divides the overall video processing workload and assigns portions to different edge nodes, allowing the system to maintain high processing efficiency through parallelization while also improving reliability by eliminating single points of failure.
Solution Approach 2:
The system combines the processing capabilities of multiple edge nodes into a unified video processing infrastructure. By merging resources across nodes and coordinating them through the orchestrator, the system achieves both high processing efficiency (through aggregated capacity) and high reliability (through redundancy and load distribution).
3Reliability
If dynamic resource rebalancing is implemented across distributed edge nodes, then video frame delivery reliability is improved, but system complexity increases
Solution Approach 1:
The patent introduces an orchestrator as an intermediary component that manages the complexity of dynamic resource rebalancing. The orchestrator acts as a central coordinator that receives task allocation requests, monitors edge node status, and makes rebalancing decisions. This intermediary absorbs the orchestration complexity, allowing individual edge nodes to remain relatively simple while the system as a whole achieves high reliability through coordinated resource management.
4Reliability
If edge nodes are over-provisioned to handle peak workloads, then service level agreement compliance is improved, but total cost of ownership increases
Solution Approach 1:
The patent implements a multi-functional resource pool where edge nodes can dynamically serve different video processing workloads based on demand. Instead of dedicating specific resources to specific tasks, the system makes resources universal and reusable across multiple functions and workloads. This allows the system to meet peak demand requirements while maintaining lower average resource levels, improving SLA compliance without proportionally increasing total resource quantity.
Data Source
AI summary
In some embodiments, infrastructure data and service data is received for a computing infrastructure. The infrastructure data indicates resources in the computing infrastructure, and the service data indicates services to be orchestrated across the computing infrastructure. An infrastructure capacity model is generated, which indicates a capacity of the computing infrastructure over a particular time window. Service-to-resource placement options are also identified, which indicate possible placements of the services across the resources over the particular time window. Resource inventory data is obtained, which indicates an inventory of resources that are available to add to the computing infrastructure during the particular time window. An infrastructure capacity plan is then generated, which indicates resource capacity allocation options over the time slots of the particular time window. Resource capacities for the services are then allocated in the computing infrastructure.