Self-Configuring Cell Computers for Seamless Storage Expansion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing storage network solutions face challenges in scalability and maintenance, as adding CPUs or disks requires service downtime and manual intervention, and maintaining data redundancy is complex, especially in distributed computing systems.
Innovation Solution
A distributed computing system utilizing a network of cell computers that can automatically configure and self-assign roles, distribute tasks, and maintain data redundancy through a social cloud architecture with a mapping technique that allows for infinite expansion and fault tolerance, using software to manage hardware failures and redundancy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If a CPU is added to speed up service execution, then processing power is improved, but service must be stopped temporarily
Solution Approach 1:
The system segments the storage and processing functions into separate components: storage cells handle data storage while compute nodes handle processing tasks. This segmentation allows independent scaling and maintenance of each component without affecting the other, enabling CPU addition without service interruption.
Solution Approach 2:
The patent introduces a control node as an intermediary that manages task distribution and coordination between storage cells and compute nodes. This intermediary enables seamless integration of new CPUs into the system without requiring service shutdown, as the control node manages the transition and load balancing.
2Quantity of substance
If a disk is added to the array, then storage capacity is improved, but the array must be reformed and services stopped
Solution Approach 1:
The storage system is segmented into independent storage cells that can be individually added or removed. Each storage cell operates autonomously with its own data chunks, allowing capacity expansion without reformatting the entire array or interrupting services.
Solution Approach 2:
The system performs preliminary data distribution and redundancy setup before adding new storage capacity. The control node pre-configures data chunk placement and redundancy relationships, enabling seamless integration of new disks without service interruption or array reformation.
3Reliability
If maintenance operations are performed, then system reliability is improved, but services are stopped or degraded
Solution Approach 1:
The system implements dynamic task migration and load balancing during maintenance operations. When a storage cell or compute node requires maintenance, the control node dynamically redistributes tasks and data to remaining healthy nodes, maintaining service availability while enabling maintenance activities.
Solution Approach 2:
The system temporarily discards tasks or data from nodes undergoing maintenance and recovers them after maintenance completion. The control node manages this process by redistributing affected tasks to other nodes during maintenance and automatically recovering them post-maintenance without service interruption.
4Reliability
If data redundancy is maintained through traditional RAID, then data integrity is improved, but system complexity and manual intervention requirements increase
Solution Approach 1:
The system implements self-service automation through the control node, which automatically manages data redundancy, distributes data chunks across storage cells, and handles failure recovery. This eliminates the need for manual RAID configuration and intervention, reducing system complexity while maintaining data integrity through automated redundancy management.
Data Source
Figure 1~2
Figure 3~4
Figure 5
AI summary
A computing system has a plurality of computers connected via a computer network to form a computing entity. Each of the computers operates substantially independent of others and configured to: interrogate network infrastructure of the computer network to determine the identity of the computing entity when the computer is connected to the computer network and thus join the computing entity by announcing its presence in the computing entity; determine an identifier of the computer in the computing entity based on the connectivity configuration in the network infrastructure and assume a role to perform a portion of operations of a computing request directed to the computing entity over the computer network, based on the presence data of the computers in the entity; identify portions of a computing job that are assigned to the computer, if any, based on identification information of the computing job and identifiers of a list of computing devices present in a computing network to process the computing job; and use a rule engine to apply a set of predetermined rules to conditions relevant to changes of presence data of computers in a computing network forming a computing entity in which data stored in the computing entity is distributed among the computers for redundancy and data recovery.