Reversible Partitioning Router for Distributed Storage Snapshots
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Distributed computing systems face inefficiencies in resource allocation and snapshot management, leading to uneven resource distribution and increased costs due to the growth of snapshots and volumes, which can result in corrupt snapshots during partitioning.
Innovation Solution
A reversible partitioning system for block-level data storage that utilizes a router with a finite state machine to manage routing metadata, determining the location for snapshot creation based on active and proposed metadata, and preventing corrupt snapshots by using hash value ranges and merit scores to optimize resource usage and partitioning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If the number of snapshots and volumes is increased to meet storage needs, then storage capacity and functionality are improved, but computing resources required to service these snapshots and volumes become excessively large and unevenly distributed
Solution Approach 1:
The patent divides the representation into multiple cells, where each cell manages a subset of snapshots and volumes. This segmentation distributes the computing resource burden across multiple nodes rather than concentrating it in a single location, enabling the system to handle large numbers of snapshots and volumes without requiring excessive total computing resources.
2Reliability
If snapshots are created to preserve state information, then data backup and recovery capabilities are improved, but the risk of creating corrupt snapshots during partitioning operations increases
Solution Approach 1:
The patent performs partitioning operations on the representation structure before creating new snapshots. By pre-partitioning the representation into multiple cells with properly configured metadata, the system ensures that subsequent snapshot creation operations do not risk creating corrupt snapshots, as the partitioning structure is already in place and validated.
3Loss of energy
If resource allocation is optimized to reduce costs, then expense and overhead are reduced, but the complexity of managing and routing requests increases
Solution Approach 1:
The patent introduces a router as an intermediary component that manages request routing between clients and cells. The router uses routing metadata to intelligently direct requests to the appropriate cell, simplifying the overall system architecture by centralizing the routing logic rather than distributing complex management responsibilities across multiple components.
Data Source
AI summary
A router receives a request to access a resource. In response to the request, the router identifies routing metadata comprising first metadata specifying a first location of the resource and a first merit value and second metadata specifying a second location of the resource and a second merit value. The router determines, based at least in part on the first merit value of the first metadata and the second merit value of the second metadata, an authoritative state of the resource associated with the first location or the second location. Based at least in part on the determination, the router identifies a preferred location for a new resource. In the event of a rollback, merit values are assigned to cause a reversal of the preferred path for the creation of new resources.