Popularity-Based Load Balancing for Fog-Cloud Placement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems are inefficient in determining which requests should be computed in the Fog and which should be forwarded to the Cloud, leading to suboptimal resource utilization and increased costs due to sudden bursts of requests.
Innovation Solution
A popularity-based load-balancing approach where a load-balancer determines the popularity of network-based resources and forwards popular requests to the Fog network while forwarding less popular or overload requests to the Cloud, using a combination of count-based popularity calculations and a Least-Recently-Used (LRU) filter to optimize cache hits and minimize latency and costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If requests are forwarded to the Fog network for processing, then latency is reduced and response time is improved, but the Fog network capacity is limited and cannot handle sudden bursts of requests
Solution Approach 1:
The system dynamically adjusts the routing of requests between Fog and Cloud based on real-time popularity metrics. The load balancer continuously monitors request patterns and adapts the forwarding decisions, switching from static Fog processing to dynamic routing that can escalate to Cloud when Fog capacity is exceeded during request bursts.
Solution Approach 2:
The invention changes the parameter of request routing by using popularity-based metrics to determine whether requests should be forwarded to Fog or Cloud. By monitoring popularity thresholds and system capacity, the load balancer changes the routing parameter dynamically, directing popular requests to Fog for low latency while routing less popular or excessive requests to Cloud to prevent overload.
2Productivity
If requests are forwarded to the Cloud network for processing, then the system can handle sudden bursts of requests, but latency increases and response time is slower
Solution Approach 1:
The system applies local quality by routing different types of requests to different networks based on their characteristics. Popular requests that can tolerate slightly higher latency are routed to Cloud for capacity, while unpopular or time-sensitive requests are handled locally at Fog to minimize latency. This creates quality differentiation in request handling based on popularity and urgency.
3Productivity
If popular requests are forwarded to the Fog network, then cache hits increase and efficiency is improved, but the Fog network may become overloaded during request bursts
Solution Approach 1:
The load balancer implements feedback by continuously monitoring Fog network load and popularity metrics. When the Fog network approaches capacity during request bursts, the system receives feedback about the overload condition and adjusts routing decisions accordingly, redirecting excess requests to Cloud to maintain stability while still maximizing cache hits for popular requests within Fog capacity.
4Ease of operation
If all requests are forwarded to the Fog network regardless of popularity, then the system simplifies routing logic, but resource utilization becomes inefficient and costs increase
Solution Approach 1:
The system performs preliminary action by pre-calculating and maintaining popularity metrics for requests before making routing decisions. The load balancer proactively identifies popular requests through continuous monitoring and prepares routing decisions in advance, so when requests arrive, the system can efficiently direct them to Fog or Cloud based on pre-computed popularity data, maintaining simplicity while improving efficiency.
Data Source
AI summary
In one embodiment, a load-balancer in a computer network receives an end-device request for a particular network-based resource, and determines a popularity of the particular network-based resource being requested in comparison to other network-based resources requested by end devices in the computer network. In response to the particular network-based resource being popularly requested, the load-balancer forwards the end-device request into a Fog network to cause the Fog network to respond to the end-device request with particular network-based resource. Conversely, in response to the particular network-based resource not being popularly requested, the load-balancer forwards the end-device request into a Cloud network to cause the Cloud network to respond to the end-device request with particular network-based resource.