Overlay Network Workload Placement Algorithm
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is currently no automated process for determining which resource-on-demand systems have the capacity to accept and run a new workload in grid computing services and utility-based data centers, leading to inefficiencies in workload placement.
Innovation Solution
An overlay network is used to store information about resource-on-demand systems, and a distributed control algorithm, such as an ant-based algorithm, is executed to select a suitable resource-on-demand system to host a workload by normalizing resource capacities and utilizing probability-based selection methods to ensure efficient workload placement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If no automated process is used for determining workload placement, then manual intervention is required to place workloads, but this leads to inefficiency and inability to automatically identify systems with sufficient capacity
Solution Approach 1:
The system enables automated self-service workload placement by having resource-on-demand systems automatically report their capacity information to the overlay network, and the distributed control algorithm automatically selecting appropriate systems without human intervention. The overlay network nodes autonomously execute the algorithm to place workloads based on current system capacities.
2Productivity
If distributed control algorithm is implemented for automated workload placement, then productivity and efficiency improve, but system complexity increases due to overlay network and algorithm implementation
Solution Approach 1:
The system segments the workload placement function into distributed overlay network nodes that independently execute the control algorithm. Each node maintains information about specific resource-on-demand systems and makes local placement decisions, avoiding the need for a centralized complex system while achieving automated placement across the entire network.
3Measurement precision
If overlay network stores information about multiple resource-on-demand systems, then accurate capacity identification is improved, but information management complexity increases
Solution Approach 1:
The system implements feedback mechanisms where resource-on-demand systems continuously report their capacity information to the overlay network, and the distributed control algorithm uses this feedback to dynamically adjust workload placement decisions. This ensures accurate capacity assessment while automating the information management process.
Data Source
AI summary
Each of a plurality of nodes in an overlay network includes information for a corresponding resource-on-demand system, wherein the information may be used to select a resource-on-demand system to host the workload. A request to place a workload is received at a node in an overlay network. A distributed control algorithm is executed for selecting a resource-on-demand system host the workload.