Blockchain State Snapshot Sharing for Disk Read Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing blockchain transaction processing systems face inefficiencies due to the high number of disk read operations required for updating states, which leads to performance bottlenecks and reduced disk lifespan.
Innovation Solution
A blockchain transaction processing method and device that reduces disk read operations by creating a state snapshot including the state values required for processing new blocks and sharing this snapshot among full nodes, thereby reducing the load on individual disk resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the entire blockchain state information is stored on disk through a key-value database, then the node can maintain and validate the complete state, but the frequent disk read and write operations become a performance bottleneck and reduce disk lifespan
Solution Approach 1:
The patent performs preliminary actions by pre-calculating and storing state snapshots at specific block intervals before they are needed for transaction processing. Full nodes advance-process state transitions and store snapshots in advance, so when transaction validation is needed, the data is already prepared and available in memory rather than requiring disk reads at the moment of processing.
Solution Approach 2:
The patent segments the continuous blockchain state into discrete snapshots taken at specific block intervals. Instead of treating the state as a continuous stream requiring frequent disk access, it divides the state history into manageable snapshot segments that can be loaded into memory and reused for multiple transaction validations within each segment's range.
2Productivity
If state information is cached in memory to improve performance, then transaction processing speed increases, but the cache size is limited and cannot accommodate the entire blockchain state
Solution Approach 1:
The patent creates copies of the blockchain state at specific intervals (snapshots) that can be stored in memory. These snapshot copies represent historical states at particular block heights and can be reused for validating transactions that occurred within the snapshot's time range, eliminating the need to load the entire current state into memory for every validation operation.
3Productivity
If new data structures are used to improve Merkle tree efficiency, then read/write overhead may be reduced, but compatibility issues arise with existing blockchain data accumulated over years
Solution Approach 1:
Instead of changing the underlying Merkle tree data structure which would cause compatibility issues, the patent performs preliminary actions by pre-computing state snapshots using the existing Merkle tree structure. This allows the system to optimize transaction validation by having state data ready in advance without modifying the core data structure that maintains compatibility with existing blockchain protocols and accumulated data.
Data Source
AI summary
The present disclosure provides a method and apparatus for processing blockchain transactions that may reduce a number of disk read operations in a full node in a blockchain network. The method includes: reading a state value required for processing a transaction of a new block from a local disk in a same order as an order of processing transactions of a plurality of new blocks; creating a state snapshot including the state value; and sharing the state snapshot with at least one second full node.