Multi-grid computing mechanism using peer-to-peer protocols
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional grid computing systems are monolithic, leading to performance bottlenecks and single points of failure due to reliance on a single master node, which limits scalability and flexibility in resource utilization.
Innovation Solution
Implementing a multi-grid mechanism using an autonomic grid computing mechanism and peer-to-peer platform protocols, allowing multiple master nodes and grid cells within a single network, enabling dynamic job submission and resource allocation based on criteria such as speed and resource requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single master node is used in conventional grid architecture, then the system structure is simple, but performance bottlenecks and single points of failure occur
Solution Approach 1:
The patent divides the monolithic grid into multiple independent grid cells, each with its own master node. This segmentation eliminates the single point of failure by distributing control functions across multiple nodes. Each grid cell operates semi-autonomously, allowing continued operation even if one master node fails.
2Ease of operation
If a single master node is used, then configuration and management are easier, but scalability is limited
Solution Approach 1:
By segmenting the grid into multiple grid cells with independent master nodes, the system achieves scalability while maintaining ease of operation. Each grid cell can be independently configured and managed, allowing the system to grow by adding more grid cells without overwhelming complexity in the central management structure.
3Adaptability or versatility
If multiple master nodes are implemented, then scalability and reliability improve, but system complexity increases
Solution Approach 1:
The segmentation into grid cells with independent master nodes enables scalability without proportionally increasing overall system complexity. Each master node manages only its local grid cell, reducing the complexity burden on any single node while maintaining the ability to scale the entire system by adding more grid cells.
Solution Approach 2:
The patent introduces an intermediary layer between job submitters and compute nodes, where grid cells act as mediators that handle job distribution and resource allocation. This intermediary structure simplifies the system architecture by localizing control functions and reducing direct complexity between distant components.
4Ease of manufacture
If manual configuration is used, then initial setup is straightforward, but deployment time increases
Solution Approach 1:
The patent implements self-service mechanisms where grid cells and compute nodes automatically discover each other and configure themselves using peer-to-peer protocols. This eliminates the need for extensive manual configuration while enabling rapid deployment, as nodes autonomously integrate into the grid without human intervention for each connection.
Data Source
AI summary
A multi-grid mechanism using peer-to-peer (P2P) platform protocols. An autonomic grid computing mechanism may be used to implement a multi-grid grid computing system. In the multi-grid grid computing system, there may be multiple master nodes, and thus multiple smaller grids (or grid cells) each including one or more compute nodes, in a single (larger) grid. A job submitter node may use one or more P2P platform protocols to discover the master nodes and to obtain information about the grid cells, including information that may be used by the job submitter to automatically select one of the grid cells that is suitable to execute a particular job. After selecting the grid cell, the job submitter may submit the particular job to the master node of the grid cell using the P2P platform protocols.