A hierarchical storage system and method for permissionless blockchains

Abstract

This invention discloses a hierarchical storage system and method for permissionless blockchains. The system includes: a hot node layer for storing hot data of the blockchain; a cold node layer for storing both hot and cold data of the blockchain; and an on-chain smart contract, including: a dynamic data management module for periodically mapping cold data blocks to the cold nodes for long-term storage using a consistent hashing algorithm; a storage proof verification module for verifying the continuous storage status of the cold data blocks by the cold nodes and obtaining a verification result; and an incentive settlement module for automatically executing economic incentives on the cold nodes based on the verification result and feeding back the results of the economic incentives to the dynamic data management module. This invention addresses the storage bloat problem of permissionless blockchains at the protocol level, significantly reducing storage costs, promoting decentralization, and improving the security and scalability of permissionless blockchain networks.

Description

Technical Field

[0001] This invention relates to the field of blockchain technology, and in particular to a hierarchical storage system and method for permissionless blockchains. Background Technology

[0002] As the cornerstone of Web3 and Decentralized Finance (DeFi), permissionless blockchains have experienced explosive growth in their ecosystem, leading to a dramatic expansion of on-chain data and triggering a severe storage bloat problem.

[0003] Current technologies generally employ a full-replication architecture, requiring each full node to store the complete history of transactions and state. This model, demanding that every participating full node store the entire history of transactions and state, results in high hardware costs—for example, an Ethereum full node already exceeds 1.3TB of data. This high barrier to entry forces node operations to be concentrated in the hands of a few well-funded entities, leading to de facto centralization. This contradicts the core principles of blockchain and poses a threat to the long-term security and censorship resistance of the network. Summary of the Invention

[0004] This invention provides a hierarchical storage system and method for permissionless blockchains, which addresses the shortcomings of storage bloat in existing technologies, and achieves a significant reduction in storage overhead, a substantial decrease in node entry barriers, and an effective improvement in network decentralization.

[0005] This invention provides a hierarchical storage system for permissionless blockchains, the system comprising:

[0006] The hot node layer, consisting of multiple hot nodes, is used to store the hot data of the blockchain;

[0007] The cold node layer, consisting of multiple cold nodes, is used to store both cold and hot data in the blockchain.

[0008] On-chain smart contracts include:

[0009] The dynamic data management module is used to periodically divide cold data into blocks and map them to the cold nodes for long-term storage using a consistent hashing algorithm; wherein, the cold data blocks are obtained by encoding the cold data;

[0010] The storage proof verification module is used to verify the continuous storage status of the cold node for the cold data block and obtain the verification result;

[0011] The incentive settlement module is used to automatically implement economic incentives for the cold nodes based on the verification results, and to feed back the results of the economic incentives to the dynamic data management module.

[0012] The hot data includes blocks from the most recent W periods and current state data; the cold data includes all blocks from the previous W periods, where each period consists of a fixed number of consecutive blocks; each cold node is represented by a tuple consisting of a unique identifier and locked stake.

[0013] According to the hierarchical storage system for permissionless blockchains provided by the present invention, the dynamic data management module includes:

[0014] Member management unit, used to determine the set of active cold nodes in each cycle;

[0015] A data allocation unit is used to allocate the cold data blocks to the cold nodes in the cold node set based on a weighted virtual hash ring.

[0016] The data handover unit is used to update the ownership of the cold data block when the cold node joins or leaves.

[0017] This invention also provides a hierarchical storage method for permissionless blockchains, comprising:

[0018] In response to each periodic boundary block, the dynamic data management module divides cold data into blocks and maps them to the cold nodes for long-term storage using a consistent hashing algorithm.

[0019] The storage proof verification module verifies the continuous storage status of the cold node for the cold data block and obtains the verification result.

[0020] Based on the verification results, the incentive settlement module automatically applies economic incentives to the cold nodes and feeds back the results of the economic incentives to the dynamic data management module.

[0021] According to the hierarchical storage method for permissionless blockchain provided by the present invention, the step of mapping cold data blocks to cold nodes for long-term storage using a consistent hashing algorithm specifically includes:

[0022] Each of the aforementioned cold data blocks is mapped to a logical slot in the virtual hash ring via a hash function;

[0023] Based on the locked stake of the cold node, a virtual node is allocated on the virtual hash ring for each cold node, wherein the number of virtual nodes is proportional to the locked stake;

[0024] The virtual nodes are placed at equal intervals on the virtual hash ring;

[0025] The logical slot is assigned to the first virtual node encountered in a clockwise direction on the virtual hash ring, and the cold node that owns the virtual node stores the cold data block.

[0026] According to a hierarchical storage method for permissionless blockchains provided by the present invention, the cold data blocks are obtained by encoding the cold data, including:

[0027] The cold data is encoded into a canonical set consisting of k+m cold data blocks using Reed-Solomon RS(k,m) erasure coding; where k is the number of original data blocks and m is the number of redundant blocks.

[0028] According to a hierarchical storage method for permissionless blockchains provided by the present invention, based on the verification results, economic incentives are automatically applied to the cold nodes, and the results of the economic incentives are fed back to the dynamic data management module, specifically including:

[0029] Based on the verification results, the proven storage share of the cold node is determined;

[0030] Based on the locked rights of the cold node, determine the virtual node weight share of the cold node;

[0031] Based on the weight share of the virtual nodes, basic rewards are distributed to the cold nodes that participate honestly.

[0032] Based on the proven storage share, service fee rewards are distributed to the cold nodes that participate honestly.

[0033] Based on the duration of failure to submit valid storage proof, a progressive penalty is imposed on the cold node, and feedback is sent to the dynamic data management module. When the duration exceeds the maximum preset tolerance threshold, the cold node is forcibly removed.

[0034] According to a hierarchical storage method for permissionless blockchains provided by the present invention, verifying the continuous storage status of the cold node for the cold data block specifically includes:

[0035] At the start of the cycle, a globally unique and unpredictable initial challenge is generated based on a verifiable random function;

[0036] Based on the initial challenge, each cold node sequentially generates a storage proof by sampling the cold data blocks in multiple rounds within a period, and forms a time-locked evidence chain based on the storage proof through a verifiable delay function;

[0037] At the end of the cycle, a converged proof is formed based on the aforementioned chain of evidence;

[0038] The on-chain smart contract performs a single verification of the aggregated proof.

[0039] According to a hierarchical storage method for permissionless blockchains provided by the present invention, after the cold data is stored in the cold node, the method for retrieving the cold data further includes:

[0040] Obtain the set of active cold nodes;

[0041] Based on the set of cold nodes, multiple cold nodes are determined to store the cold data blocks to be retrieved;

[0042] Send data requests to multiple cold nodes in parallel;

[0043] When each cold data block arrives, verify the integrity of the cold data block; iterate until the number of cold data blocks reaches the original number of data blocks;

[0044] Based on all the aforementioned cold data blocks, the cold data is reconstructed through decoding.

[0045] According to the present invention, a hierarchical storage method for permissionless blockchains is provided, wherein the storage proof is a storage time proof.

[0046] According to a hierarchical storage method for permissionless blockchains provided by the present invention, the step of placing the virtual nodes at equal intervals on the virtual hash ring further includes:

[0047] The location of the first virtual node is determined by the unique identifier of the cold node that owns the virtual node.

[0048] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the hierarchical storage method for permissionless blockchains as described above.

[0049] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements a hierarchical storage method for permissionless blockchains as described above.

[0050] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention decouples the storage overhead of blockchain from the network scale by layering hot and cold data and storing hot and cold nodes. Combined with the closed-loop accountability framework formed by the dynamic data management module, storage proof verification module and incentive settlement module integrated in on-chain smart contracts, this invention solves the storage bloat problem for permissionless blockchains at the protocol level, significantly reduces storage costs, promotes decentralization, and improves the security and scalability of permissionless blockchain networks. Attached Figure Description

[0051] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0052] Figure 1 This is a schematic diagram of the architecture of HC-Store, a hierarchical storage system for permissionless blockchains according to an embodiment of the present invention.

[0053] Figure 2 This is a flowchart illustrating a hierarchical storage method for permissionless blockchains according to an embodiment of the present invention.

[0054] Figure 3 This is an example diagram of data allocation for a virtual hash ring according to an embodiment of the present invention.

[0055] Figure 4 This is a schematic diagram of data redistribution when a new node is added according to an embodiment of the present invention.

[0056] Figure 5 This is a schematic diagram of a user-driven cold data retrieval process according to an embodiment of the present invention.

[0057] Figure 6 This is a comparative schematic diagram showing the variation of single-node storage overhead with block height in various storage systems according to embodiments of the present invention.

[0058] Figure 7 This is a comparative diagram showing the variation of total network storage overhead with block height for various storage systems according to embodiments of the present invention.

[0059] Figure 8 This is a comparison chart of the single-node storage cost of various storage models under different network scales according to embodiments of the present invention.

[0060] Figure 9 This is a comparative diagram showing the change in total network overhead as the total number of network nodes varies across various storage systems according to embodiments of the present invention.

[0061] Figure 10 This is a comparative diagram showing how the query success rate of various storage systems varies with the probability of node unavailability according to embodiments of the present invention.

[0062] Figure 11 This is a comparative diagram illustrating the changes in query success rate of various storage systems according to embodiments of the present invention as the proportion of data deleted by malicious nodes changes.

[0063] Figure 12This is a schematic diagram illustrating how the probability of timely data recovery varies with the retrieval deadline in HC-Store according to an embodiment of the present invention.

[0064] Figure 13 This is a schematic diagram illustrating how the bandwidth overhead of HC-Store repairing a single block varies with the original data block number parameter k according to an embodiment of the present invention.

[0065] Figure 14 This is a schematic diagram comparing the profits of honest nodes and two types of witch attack nodes according to an embodiment of the present invention.

[0066] Figure 15 This is a comparative diagram showing the change in the probability of detecting malicious data deletion with the number of challenges under two RS erasure coding configurations according to an embodiment of the present invention, where (a) represents the RS(10,5) configuration and (b) represents the RS(10,10) configuration.

[0067] Figure 16 This is a schematic diagram illustrating how the break-even user query price varies with block height, which is required for economic sustainability according to an embodiment of the present invention.

[0068] Figure 17 This is a schematic diagram illustrating how the break-even query volume required for economic sustainability according to an embodiment of the present invention varies with block height. Detailed Implementation

[0069] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0070] Existing solutions to storage bloat, when directly applied to decentralized, trustless blockchain environments, will face a series of severe challenges stemming from the inherent properties of blockchain:

[0071] Node asymmetry: The peer-to-peer nature of blockchain emphasizes equality among nodes, while storage tiers inevitably introduce an asymmetry in node roles. Reconciling this inequality within a decentralized governance model is a major challenge.

[0072] Trustless accountability: In the absence of a trusted third party, how to verify that a dedicated subset of nodes is honestly fulfilling its storage responsibilities and how to drive these high-cost tasks through economic incentives is a core challenge in ensuring long-term stable operation.

[0073] Network dynamism: In permissionless networks, the dynamic joining and leaving of nodes poses significant challenges to hierarchical management of hot and cold data. Without effective maintenance mechanisms, this can easily lead to data loss and unbalanced storage loads.

[0074] To address the challenge of blockchain storage bloat, existing technologies have significant limitations, as shown in Table 1.

[0075] Table 1 Comparison of Blockchain Storage Architecture Dimensions

[0076]

[0077] Icon descriptions: ○: Low / None; ◐: Medium; ●: High; ✓: Natively supported; ✗: Not supported.

[0078] This invention provides a novel hierarchical storage architecture that fundamentally decouples network participation costs from the historical length of the chain through a two-layer architecture with heterogeneous hot and cold node roles, thereby eradicating the storage bloat problem. Through a unified and accountable storage framework that synergistically combines efficient proof-of-storage time (ePoSt) with a robust staking and slashing economic model, a fully accountable system is created that enforces reliable data custody without relying on trust or altruism. A set of protocols supporting dynamic and resilient operation enables the architecture to run in a dynamic, permissionless environment, including a stake-weighted virtual hash ring for efficient, Sybil-resistant data management during node churn, and a fault-tolerant parallelization mechanism for users to quickly retrieve cold data.

[0079] like Figure 1 As shown, this invention provides a hierarchical storage system for permissionless blockchains, comprising the following modules:

[0080] Hot node layer 100, consisting of multiple hot nodes, is used to store the hot data of the blockchain;

[0081] The cold node layer 110 consists of multiple cold nodes and is used to store the cold and hot data of the blockchain.

[0082] 120 on-chain smart contracts, including:

[0083] The dynamic data management module is used to periodically divide cold data into blocks and map them to the cold nodes for long-term storage using a consistent hashing algorithm; wherein, the cold data blocks are obtained by encoding the cold data;

[0084] The storage proof verification module is used to verify the continuous storage status of the cold node for the cold data block and obtain the verification result;

[0085] The incentive settlement module is used to automatically implement economic incentives for the cold nodes based on the verification results, and to feed back the results of the economic incentives to the dynamic data management module.

[0086] The hot data includes blocks from the most recent W periods and current state data; the cold data includes all blocks from the previous W periods, where each period consists of a fixed number of consecutive blocks; each cold node is represented by a tuple consisting of a unique identifier and locked stake.

[0087] Specifically, the tiered storage system for permissionless blockchains integrates two core concepts: data tiering and storage tiering. Data tiering differentiates based on the timeliness and access frequency of ledger data, while storage tiering differentiates based on the roles and responsibilities of network nodes. This two-tiered approach allows most participants to operate with minimal storage overhead, while facilitating the delegation of the long-term preservation of the complete ledger to a dedicated and accountable subset of nodes.

[0088] Therefore, the system's state transition model is built on epochs. Starting from the genesis block, a fixed number of consecutive blocks constitute an epoch, where the first block is a special epoch boundary block (EBB). The EBB serves as the synchronization point for the entire network. Based on this timeline, the system's data hierarchy is defined as follows: Hot data includes the most recent... The data includes blocks within each period and complete current state data; cold data includes this window (the most recent block). All historical blocks prior to (a certain number of cycles). The hierarchical storage architecture reduces storage overhead from O(N·H) of existing technologies to O(H) by separating hot data from cold data, where N is the number of nodes and H is the length of history. This decouples storage costs from network scale, significantly reducing storage costs and enabling the blockchain system to support large-scale networks and long-term historical data storage, thereby enhancing system scalability.

[0089] The storage tiers of the HC-Store system in this embodiment of the invention are coordinated by on-chain smart contracts 120 and involve multiple participants. For example, users query data from network nodes. Some nodes, acting as miners, are responsible for proposing new blocks, including special EBBs that trigger critical system protocols. In terms of storage management, nodes are divided into two main roles: most nodes act as lightweight hot nodes, forming a high-performance layer that stores only hot data; and a few dedicated cold nodes form a distributed archiving layer that stores all hot data, plus a copy of archived cold data allocated according to deterministic rules.

[0090] To address the loss of cold nodes in permissionless blockchain environments, this invention provides a dynamic data management module responsible for running at each EBB (Entity Blockchain Base) to maintain data health and load balancing. The basic data flow follows two paths. First, during the ongoing hot data replication process, each new block is initially hot data and replicated across all nodes. Second, during the periodic data state transition and allocation process triggered by each EBB, the oldest hot data cycle is reclassified as cold data. This new cold data unit can be archived through two automated steps: 1) using Reed-Solomon... The encoding scheme encodes it into a single... 1) A canonical set consisting of several blocks; 2) Each block is deterministically assigned to a specific cold node for long-term storage through a mapping function based on a consistent hashing algorithm. The consistent hashing algorithm can be implemented through a virtual hash ring, a distributed hash table, or any other algorithm that satisfies the characteristics of load balancing, scalability, and monotonicity, without any restrictions. The cold data blocks are obtained by encoding the cold data; the encoding method can be Reed-Solomon, Cauchy-Reed-Solomon, or any other encoding method that can divide the data into blocks and provide redundancy and fault tolerance. Through the dynamic data management module, in a network where cold nodes can dynamically join and leave, the system ensures the long-term availability of cold data while maintaining a fair and efficient allocation of storage load.

[0091] Continuous storage state refers to the state in which a cold node continuously and uninterruptedly holds and accesses its allocated cold data blocks within a specific time period. After the dynamic data management module completes the storage of cold data blocks, the storage proof verification module verifies, based on the continuous storage state, whether the cold node has continuously and correctly stored its allocated cold data blocks (hereinafter referred to as blocks) using the efficient proof of storage time (ePoSt) protocol, thus providing an objective basis for economic incentives and penalties. This module ensures the stability and reliability of the storage provider set. Incentive compatibility is achieved by making diligent participation the most profitable strategy for rational nodes. The storage proof verification module can enforce data hosting, ensuring that cold nodes continuously store the correct allocated data under verifiable conditions.

[0092] After obtaining the verification result of the storage proof verification, the incentive settlement module automatically applies economic incentives to cold nodes, distributing rewards to nodes that participate honestly, and feeding back the results of these economic incentives to the dynamic data management module. Economic incentives can include basic rewards issued by the blockchain network protocol or service fee rewards paid by users. The result of the economic incentives refers to the dynamic changes in the state of cold nodes. These changes are fed back to the dynamic data management module to update the membership, locked stake, and data allocation of cold nodes, thereby achieving system self-regulation and continuous operation. For example, a cold node that successfully receives a reward remains active; a node that fails verification or is penalized may be marked as abnormal or pending removal. Furthermore, the rewards a node receives may be automatically staked, increasing its locked stake; a cold node that seriously violates regulations may be completely penalized, with its locked stake reset to zero. This economic model reliably incentivizes nodes to assume the specialized and resource-intensive role of cold nodes.

[0093] The closed-loop accountability framework, consisting of a dynamic data management module, a storage proof verification module, and an incentive settlement module integrated into the on-chain smart contract, eliminates the need for trust, enables the system to self-regulate and operate continuously, ensures that data is always in a usable and complete state, promotes decentralization, and enhances the security of the blockchain network.

[0094] This invention decouples the storage overhead of blockchain from network scale by layering hot and cold data and hot and cold nodes. Combined with a closed-loop accountable framework consisting of a dynamic data management module, a storage proof verification module, and an incentive settlement module integrated into on-chain smart contracts, this invention solves the storage bloat problem for permissionless blockchains at the protocol level, significantly reduces storage costs, promotes decentralization, and improves the security and scalability of permissionless blockchain networks.

[0095] Optionally, the dynamic data management module includes:

[0096] Member management unit, used to determine the set of active cold nodes in each cycle;

[0097] A data allocation unit is used to allocate the cold data blocks to the cold nodes in the cold node set based on a weighted virtual hash ring.

[0098] The data handover unit is used to update the ownership of the cold data block when the cold node joins or leaves.

[0099] Specifically, an on-chain member contract will eventually determine the set of active cold nodes in each cycle. It processes (i) registration transactions containing locked stake, (ii) voluntary deregistration requests, and (iii) nodes forcibly removed due to violations of liveness requirements (e.g., consecutive ePoSt failures). For each active cold node, the contract records a tuple consisting of its unique identifier and locked stake. This set of authority and its associated equity values ​​are the sole input for data allocation, rebalancing, and repair within this period.

[0100] To achieve dynamic, efficient, and fair block distribution, HC-Store allocates cold data blocks on a stake-weighted virtual hash ring. This achieves stake-based load balancing by mapping physical nodes to multiple virtual nodes, allowing high-performance nodes to handle more storage tasks and low-performance nodes to handle fewer. A node's influence is proportional to its stake; dispersed stake prevents the generation of more benefits, thus fundamentally resisting Sybil attacks.

[0101] The following describes the hierarchical storage method for permissionless blockchains provided by the present invention. The hierarchical storage method for permissionless blockchains described below can be referred to in correspondence with the hierarchical storage system for permissionless blockchains described above.

[0102] This invention also provides a hierarchical storage method for permissionless blockchains, comprising the following steps:

[0103] Step 200: In response to each periodic boundary block, the cold data blocks are mapped to the cold nodes for long-term storage through the dynamic data management module using a consistent hashing algorithm.

[0104] Step 201: Verify the continuous storage status of the cold node for the cold data block through the storage proof verification module, and obtain the verification result.

[0105] Step 202: Based on the verification results, the incentive settlement module automatically applies economic incentives to the cold node and feeds back the results of the economic incentives to the dynamic data management module.

[0106] The above steps refer to the hierarchical storage system for permissionless blockchains described above, and will not be repeated here.

[0107] Optionally, the step of mapping cold data blocks to cold nodes for long-term storage using a consistent hashing algorithm specifically includes:

[0108] Each of the aforementioned cold data blocks is mapped to a logical slot in the virtual hash ring via a hash function;

[0109] Based on the locked stake of the cold node, a virtual node is allocated on the virtual hash ring for each cold node, wherein the number of virtual nodes is proportional to the stake value;

[0110] The virtual nodes are placed at equal intervals on the virtual hash ring;

[0111] The logical slot is assigned to the first virtual node encountered in a clockwise direction on the virtual hash ring, and the cold node that owns the virtual node stores the cold data block.

[0112] Specifically, HC-Store distributes cold data blocks on a stake-weighted virtual hash ring:

[0113] (i) Permanent logical slots. Each cold data block is deterministically and permanently mapped to a static logical slot through a consistent hash function. The logical slot on the address ring, the address of the logical slot It can be obtained through the following formula:

[0114] (1)

[0115] in, Represents a hash function; It is the block identifier for the block; It is the logical sequence number of the block.

[0116] (ii) Rights correspond to virtual nodes. For each person holding rights... Active cold nodes The number of virtual nodes (vnodes) it controls It is proportional to its rights and interests:

[0117] (2)

[0118] in, It is a system parameter that represents the amount of equity required for each virtual node.

[0119] (iii) Placement of virtual nodes. Optionally, the position of the first virtual node is determined by the unique identifier of the cold node:

[0120] (3)

[0121] in, This indicates the position of the 0th virtual node of the xth cold node; The unique identifier representing the x-th cold node. The rest... The virtual nodes are placed at equal intervals on the ring to ensure that the rights and interests of the nodes are fairly reflected:

[0122]

[0123] (iv) Ownership Rules. A logical slot is owned by the first active virtual node encountered clockwise on the ring; storage responsibility is assumed by the physical cold node that controls that virtual node. This allocation method is as follows: Figure 3 As shown, two physical nodes are displayed. and Each node controls two virtual nodes (respectively). and Data partitioning (e.g.) and The logical slot addresses of the first and second blocks of block 1 are assigned to the nearest virtual node in a clockwise direction. Therefore, by physical nodes storage.

[0124] 3) Data management when a node is added

[0125] When a new cold node x is added, its A virtual node is inserted into the ring. It will then inherit the slots now located between each new virtual node and its clockwise successor virtual node. For example... Figure 4 As shown, when a virtual node is controlled and new node When a (red) node is added, it inherits the chunk from its successor node. For example, a virtual node. From its successor node Take over the division Ownership; this change only affects the local neighborhood of the inserted virtual node, ensuring the locality of the change. HC-Store coordinates a secure chunk handoff protocol for each affected slot range: (1) Copying and proof - the former owner transfers the required chunk along with the Merkle proof to the new member; (2) Ownership update - the member contract records the new owner after verification; (3) Cleanup - the former owner can delete the handed-off chunk after a grace period.

[0126] 4) Data management when a node exits

[0127] Unexpected Exit. When a cold node unexpectedly exits, all slots previously held by its virtual nodes are immediately taken over by their respective clockwise successors. For example, if a node... Upon leaving, storage allocation will be from Figure 4 The state shown has been restored to Figure 3 The protocol issues a permissionless repair task for each affected block: any "repairman" can acquire k sibling blocks (when the original data block size is k), reconstruct the lost block, and deliver it to the new owner. Successful repairs are rewarded, with the cost covered by slashing the equity of the exiting node.

[0128] Normal exit. A normally deregistered node must, within a grace window, follow the same handover protocol and push all its blocks to the physical owner of the successor virtual node. After on-chain confirmation, the node will... Remove it from the list and you can reclaim your rights.

[0129] Optionally, the cold data blocks are obtained by encoding the cold data, including:

[0130] The cold data is encoded into a canonical set consisting of k+m cold data blocks using Reed-Solomon RS(k,m) erasure coding; where k is the number of original data blocks and m is the number of redundant blocks.

[0131] Specifically, the RS(k,m) erasure code can tolerate the loss of up to m blocks while still allowing data recovery; the Maximum Distance Divisible Code (MDS) provides maximum fault tolerance with minimal redundancy; and the parameters k and m are adjustable, allowing for a balance between storage overhead and reliability as needed. This erasure code provides maximum fault tolerance with minimal storage overhead, which is beneficial for building reliable distributed storage systems.

[0132] Optionally, based on the verification results, economic incentives are automatically applied to the cold nodes, and the results of the economic incentives are fed back to the dynamic data management module, specifically including:

[0133] Based on the verification results, the proven storage share of the cold node is determined;

[0134] Based on the locked rights of the cold node, determine the virtual node weight share of the cold node;

[0135] Based on the weight share of the virtual nodes, basic rewards are distributed to the cold nodes that participate honestly.

[0136] Based on the proven storage share, service fee rewards are distributed to the cold nodes that participate honestly.

[0137] Based on the duration of failure to submit valid storage proof, a progressive penalty is imposed on the cold node, and feedback is sent to the dynamic data management module. When the duration exceeds the maximum preset tolerance threshold, the cold node is forcibly removed.

[0138] Specifically, HC-Store employs a robust economic framework corresponding to the incentive settlement process to ensure the stability and reliability of the storage provider pool. The protocol aims to achieve incentive compatibility by making diligent participation the most profitable strategy for rational nodes. It consists of three key components: staking as a prerequisite, a dual-source reward system, and a progressive penalty mechanism.

[0139] 1) Pledge as a prerequisite for participation

[0140] To become a cold node, participants must first register through on-chain transactions and lock a certain amount of network-native tokens as financial collateral (i.e., locked equity, or equity for short) in a dedicated smart contract. The amount of equity is proportional to the storage capacity committed by a node, and is determined by the number of its virtual nodes. To determine its expected load. Exit requires a period of time. There is a periodic unbinding window; the rights can only be withdrawn after all handover and repair obligations are finally completed on the chain.

[0141] 2) Periodic-based reward and punishment decisions

[0142] At each EBB (cycle i), the settlement contract uses the verified ePoSt results of each node to determine its rewards and penalties for that cycle.

[0143] (i) Dual-source rewards. In cycle i, the rewards for honest participation come from two sources: (1) the base reward from protocol issuance (inflation). (1) Rewards for nodes providing infrastructure; (2) Service fee rewards from user payments. This is used to reward nodes based on the amount of data they can provably store. The total reward pool is... .

[0144] (ii) Reward distribution. Payments are made to cold nodes. The reward is calculated based on the contributor's contribution and depends on whether they successfully submitted an ePoSt (by...). (Representation). Let... This is a set of active but cold nodes. Basic reward. Based on the virtual node weight share of cold nodes Distribute the funds, and the service fee Then based on its proven storage share Distribute:

[0145] (5)

[0146] in yes Total number of bytes of cold data successfully verified. Total payment amount:

[0147] (6)

[0148] (iii) Supplementary Incentives. To ensure data health, the agreement also provides bounties for repairers and collaborating handover senders, funded by a small portion of forfeited equity or a reward pool.

[0149] (iv) Progressive penalties. To deter negligence and malicious behavior, the protocol employs a progressive penalty mechanism based on the number of consecutive ePoSt failures of a node.

[0150] First failure: Failure to submit valid proof within one cycle ( Cold nodes will lose all potential rewards for that period, but their stake will remain unaffected.

[0151] Consecutive failures: If failures persist for a predefined short consecutive period (e.g., 2 to 1000 cycles). If it fails, a slight scrambling penalty will be imposed, removing a small portion of the node's equity in each failed cycle.

[0152] Continuous failures: Exceeding the maximum tolerance threshold Consecutive failures are interpreted as permanent departure or malicious behavior. This will trigger severe severance penalties, automatically confiscating all stake in the node and forcibly removing it from the active cold node set. Removed from the middle.

[0153] Dual-source incentives ensure the economic sustainability of the system, while gradual penalties ensure its security and stability. The combination of these two approaches constructs an incentive-compatible economic model.

[0154] Optionally, verifying the persistent storage status of the cold node for the cold data block specifically includes:

[0155] At the start of the cycle, a globally unique and unpredictable initial challenge is generated based on a verifiable random function;

[0156] Based on the initial challenge, each cold node sequentially generates a storage proof by sampling the cold data blocks in multiple rounds within a period, and forms a time-locked evidence chain based on the storage proof through a verifiable delay function;

[0157] At the end of the cycle, a converged proof is formed based on the aforementioned chain of evidence;

[0158] The on-chain smart contract performs a single verification of the aggregated proof.

[0159] Specifically, proof of storage refers to the efficient storage time proof, ePoSt. HC-Store implements its proof of storage verification mechanism through this protocol. This protocol is tightly coupled with the system architecture, providing strong, non-interactive cryptographic guarantees for continuous data hosting. Its end-to-end process within a single HC-Store cycle enforces the economic model and includes four key phases:

[0160] 1) Challenge Start

[0161] At the start of period i, the system uses a Verifiable Random Function (VRF) to deterministically generate a globally unique and unpredictable initial challenge from the hash value of the period boundary block (EBB). This non-interactive approach ensures consistency across the entire network of challenges and makes them unpredictable in advance, thus eliminating the need for centralized validators.

[0162] 2) Sequential proof generation

[0163] A predetermined quantity within a period In each round, each cold node must sequentially compute the proof during a multi-round sampling process. In the j-th round, the node uses the current challenge. A specific cold data block is sampled from its allocated set, a Proof of Storage (PoS) is computed for it, and then a Verifiable Delay Function (VDF) is executed on the PoS result. The output of the VDF is used to derive the next challenge. This forms a time-locked chain of evidence, cryptographically proving the passage of time and the continued possession of data.

[0164] 3) Prove Aggregation and Submission

[0165] At the end of the cycle, to maximize on-chain efficiency, nodes utilize zero-knowledge succinct non-interactive argument of knowledge (zk-SNARKs) to prove all... The proofs of the rounds are compressed and aggregated into a single, constant-size aggregate proof. This aggregated proof verifies the correct execution of the entire sequential proof process and is then submitted to the on-chain smart contract as a single transaction.

[0166] 4) On-chain verification

[0167] Smart contracts only require aggregate proofs Perform a single, efficient verification. Compared to verifying a single proof, this design reduces on-chain verification costs by several orders of magnitude, making continuous, large-scale storage audits economically feasible. A successful verification provides a cryptographic guarantee that the node correctly executed the entire sequential, time-locked sampling process. This binary result directly informs the incentive settlement protocol, which then distributes the appropriate reward or penalty.

[0168] Through the above approach, HC-Store can effectively verify and ensure the continuous availability of data in a decentralized environment. This verifiability, which does not rely on a specific central authority, is key to building a robust decentralized storage network and helps improve the data reliability and user trust of the entire blockchain network.

[0169] Optionally, after the cold data is stored in the cold node, the method for retrieving the cold data further includes:

[0170] Obtain the set of active cold nodes;

[0171] Based on the set of cold nodes, multiple cold nodes are determined to store the cold data blocks to be retrieved;

[0172] Send data requests to multiple cold nodes in parallel;

[0173] When each cold data block arrives, verify the integrity of the cold data block; iterate until the number of cold data blocks reaches the original number of data blocks;

[0174] Based on the cold data blocks, which are at least the number of original data blocks, the cold data is reconstructed by decoding.

[0175] Specifically, such as Figure 5 As shown, users can efficiently retrieve archived cold data when needed. HC-Store provides an efficient, user-driven, and fault-tolerant cold data retrieval mechanism that operates without any centralized index. The entire process is initiated by the user client and follows these six key steps:

[0176] Step 1: Request Initiation and Node Discovery. The client first queries the on-chain member contract to obtain the latest set of active cold nodes. .

[0177] Step 2: Local Block Mapping. The client uses the list of active nodes to execute a virtual hash ring algorithm locally, thereby deterministically identifying the node responsible for storing the required historical data. Within each block, each block is specifically handled by a particular cold node.

[0178] Step 3, Parallel Requests. To minimize latency and provide resilience to unresponsive nodes, the client sends data requests in parallel to multiple nodes identified in the previous step.

[0179] Step 4, Integrity Verification. Upon arrival of each block, the client immediately verifies its integrity. Responding cold nodes must include a Merkle proof in the block, which the client verifies against the immutable Merkle root stored in the original block header. This prevents malicious nodes from providing corrupted data.

[0180] Step 5: First-to-Arrive Termination. The client employs a "first-to-arrive termination" strategy. It continuously receives and verifies blocks until it has collected at least k distinct valid blocks. At this point, it immediately stops waiting for any requests that have not yet arrived.

[0181] Step 6, Local Reconstruction. Once k valid blocks have been collected, the client uses... Erasure coding decoding algorithms reconstruct the complete original data block locally. This approach ensures that retrieval time is determined by the k-th fastest responding node, not the slowest, and provides inherent fault tolerance for up to m failed or unresponsive nodes.

[0182] The above method, through a parallel request mechanism and a first-to-arrive-terminate strategy, significantly reduces data retrieval latency and improves overall throughput. It inherently possesses fault tolerance, and combined with an integrity verification step (verifying Merkel proofs), effectively prevents malicious nodes from providing corrupted data. The entire retrieval process is driven by the user client, eliminating reliance on centralized index servers or trusted third parties, thus promoting decentralization and providing an efficient, reliable, secure, and cost-effective decentralized storage experience.

[0183] The above is a detailed description of the methods of the embodiments of the present invention. The following provides theoretical support for the embodiments of the present invention through system analysis and practical support for the embodiments of the present invention through experiments, further verifying the beneficial effects of the embodiments of the present invention.

[0184] (1) System Theoretical Analysis

[0185] A comprehensive, multi-dimensional analysis of the HC-Store architecture is conducted. Its core claims are validated through a series of formal theorems, which quantify the system across five key dimensions: storage efficiency, data availability, performance, security, and economic feasibility.

[0186] A. Storage efficiency analysis

[0187] A primary goal of HC-Store is to fundamentally alleviate the storage burden on network participants. Unlike traditional blockchains where storage costs scale with both network size and total number of cycles, HC-Store decouples the cost of storing cold data from the number of nodes. The following theorem formally describes this improvement.

[0188] Theorem 1 (Storage Complexity): In the HC-Store architecture, assume the block size is fixed and the number of blocks in each cycle is constant. Let... The total number of nodes. The number of cold nodes. The total number of cycles, The total network storage cost, The expected storage cost for a single cold node. In a long-running blockchain, As the dominant variable, under these typical conditions, the following complexity holds true:

[0189] (a) ;

[0190] (b)

[0191] Proof Summary: This proof is based on the analysis of thermal data (as...) (Extended) and cold data (with) The total network complexity is derived by modeling the storage cost (of the extended version). Under the conditions of the theorem, it simplifies to Total cost of cold data storage Assigned to Among the cold nodes, the expected complexity of a single node is... .

[0192] This theorem formally establishes that HC-Store breaks through the bottleneck of storage expansion. Conclusion (a) shows that the total storage cost of the system only increases with its historical length ( This is an extension, which differs from traditional architectures. The complexity differs. This is a fundamental improvement that makes the network's storage overhead independent of its size. Crucially, conclusion (b) demonstrates that this advantage also benefits individual participants. By distributing the archiving load, the expected storage complexity per cold node is reduced to... This is only what a traditional full node requires. Complexity The significant reduction in hardware requirements will lower the barrier to entry, thereby promoting the formation of a more decentralized and robust collection of storage providers.

[0193] B. Data Availability Analysis

[0194] In HC-Store, the availability of historical data is crucial. It depends not only on the algebraic properties of erasure coding but also on the system's ability to recover from both benign and malicious node failures in a permissionless environment. The following theorems establish the fundamental conditions for data recovery and the system's strong probabilistic guarantees against node unavailability.

[0195] Theorem 2 (Data Availability). In the HC-Store architecture, consider a user... Scheme coding and distribution Data blocks in a cold node. Let's assume... Let the number of its unusable blocks be denoted as . for The proportion of unavailable nodes. Let... This represents the probability of successfully recovering the data. The following properties must hold true:

[0196] (a) A data block is recoverable if and only if the number of its unavailable blocks satisfies .

[0197] (b) .

[0198] Proof Summary: (a) The proof of (a) stems directly from the algebraic properties of Reed-Solomon codes. (b) The proof of (b) will... The allocation of a block to the set of unavailable nodes is modeled as a series of Bernoulli trials, where the number of damaged blocks follows a binomial distribution. If this number does not exceed... The data is then recoverable.

[0199] This theorem comprehensively describes the availability of data. Conclusion (a) defines the absolute algebraic requirement, while conclusion (b) provides a powerful and practical tool for system design. It allows for a quantitative assessment of a system's ability to withstand failures at a specific proportion of nodes. For example, in a system composed of… In a network composed of cold nodes, even with a high number of cold nodes... The node is unavailable. Use highly redundant The data encoded by the scheme still has a recovery probability exceeding This demonstrates the system's robustness and allows designers to adjust redundancy parameters. To achieve the desired level of data availability under a specific threat model.

[0200] C. Performance Analysis

[0201] A scalable storage system must excel in two key scenarios: responding to user requests for historical data (retrieval) and efficiently maintaining data integrity during node churn (repair). The following theorem provides an analytical model for end-to-end retrieval latency and network costs of data repair.

[0202] Theorem 3 (System Performance). In the HC-Store protocol, consider using... The scheme is encoded into a size of The data is divided into blocks. Let it come from... The response delay of each of the cold nodes is an independent and identically distributed response with a rate of... Let be an exponential random variable. For download bandwidth The client's expected end-to-end retrieval time, let's assume... The total network bandwidth cost required to repair a single lost segment. The following property holds true:

[0203] (a) .

[0204] (b) .

[0205] Proof Summary: The proof of (a) models the retrieval time as the sum of network latency, data transmission time, and local decoding time, where network latency is modeled as... The first independent and identically distributed exponential variable The proof of (b) is based on the reconstruction requirements of the RS code, summing the download and upload traffic of the repairer. This theorem quantifies the actual performance of the system. Conclusion (a) provides a concrete analytical model to predict the average retrieval latency and highlights key influencing factors: the number of cold nodes ( ), their individual responsiveness ( ) and client bandwidth ( This is crucial for establishing Service Level Agreements (SLAs) and identifying performance bottlenecks. Conclusion (b) demonstrates a characteristic critical to scalability: the network cost of data repair is bounded and localized. It depends only on the erasure coding parameters. And with network size ( Or chain history ( This is unrelated. This ensures that node loss does not lead to network-wide congestion, keeping the system robust in a highly dynamic environment.

[0206] D. Security Analysis

[0207] The security analysis of HC-Store focuses on two main threats: malicious behavior by a single storage provider (such as data deletion) and attacks targeting system participation mechanisms (such as Sybil attacks). The following theorems establish the cryptographic and economic defense mechanisms of the system against these threats.

[0208] Theorem 4 (System Security). In the HC-Store architecture, let... In order to be in During the challenge, a data entry was detected that had been deleted. The probability of a malicious node in proportion. Let... For a total invested capital The expected utility of the witch attacker, let The expected utility of an honest node that invests the same amount of capital. The following security guarantees hold:

[0209] (a) .

[0210] (b) .

[0211] Proof Summary: The proof of (a) is based on probabilistic analysis, which will... The challenge is modeled as an independent Bernoulli trial. The proof of (b) is a game-theoretic analysis showing that distributing a given amount of stake across multiple identities does not provide any profit advantage over consolidating it into a single identity.

[0212] This theorem highlights the synergistic effect of cryptographic and economic mechanisms in ensuring system security. Conclusion (a) provides a precise quantitative measure of the effectiveness of auditing mechanisms. For any non-zero data deletion action, as... The increase in detection probability It will rapidly approach 1. This high probability, coupled with severe stake forfeiture penalties, makes deleting data an economically irrational strategy. Conclusion (b) demonstrates that stake-weighted virtual hash rings are a key defense. By setting high entry costs (stakes) and tying influence to that stake, the protocol eliminates the primary threat of Sybil attacks and ensures the integrity of the storage provider set.

[0213] E. Economic Model Analysis

[0214] A robust economic model for a decentralized storage system must be incentive-compatible with its participants and sustainable in the long run. The following theorem establishes these two core properties for HC-Store, proving its economic feasibility.

[0215] Theorem 5 (Economic Feasibility). In the HC-Store economic model, it is assumed that the nodes are rational and aim to maximize profits. Let... For the net profit of honest participation, let For deviation behavior (through cost savings) However, there are The probability of being detected and losing the expected profit of β proportion of equity S) is given. For period Total user costs The length of the chain's history. The following properties hold true:

[0216] (a) Incentive compatibility: If In this case, honest participation is the dominant strategy.

[0217] (b) Long-term sustainability: To ensure the system remains solvent even after a reduction in the amount of protocol issued, the following conditions must be met. .

[0218] Proof Summary: (a) The proof is based on game theory analysis, comparing the expected utility of honest and deviant strategies. (b) The proof analyzes the time-varying functions of the system's economic inflows and outflows, deriving the asymptotic growth rate necessary for user fees. This theorem formally establishes the economic robustness of the HC-Store design. Conclusion (a) shows that the system achieves this through cryptographic security (high detection probability). This is directly linked to severe financial consequences, creating a strong economic deterrent against misconduct. This makes compliance with the agreement the most profitable option. Conclusion (b) offers profound insights into the network's lifecycle: it cannot rely on inflationary rewards indefinitely. Its long-term health is fundamentally coupled to its utility and its ability to generate real user fees. This establishes a clear objective for the ecosystem: it must promote real-world application deployment and implement effective fee markets to ensure that the value generated by the network can permanently fund its own security and data persistence.

[0219] (2) Experimental evaluation

[0220] To experimentally validate the theoretical claims of the HC-Store architecture, a high-fidelity prototype was developed and a series of experiments were conducted. This section provides experimental evidence for the five core propositions analyzed in Section 4: storage efficiency, data availability, performance, security, and economic feasibility.

[0221] A. Experimental Setup

[0222] A high-fidelity prototype of the HC-Store architecture was developed using Python. The simulator models the core functionalities of hot nodes, cold nodes, and user clients. To quantify the advantages of HC-Store (HC), its performance was evaluated against two benchmark models: (1) a traditional Full Replication (FR) model and (2) a conceptual Erasure-Coded (EC) model. Experiments were driven by a real-world dataset containing all Bitcoin blocks from May 1, 2025 to July 31, 2025. The actual size of each historical block was used in the simulation. All experiments were conducted on a machine equipped with an AMD Ryzen 7 7840H CPU and an NVIDIA GeForce RTX 4060 laptop GPU. Unless otherwise stated, the simulations in this embodiment use a set of global parameters: a default network size of 255 nodes, a period size of 100 blocks, and a hot data window of 2 periods.

[0223] B. Storage efficiency verification

[0224] These experiments verify the storage complexity claim of Theorem 1. For example... Figure 6 and Figure 7 As shown, the storage cost of the FR model increases sharply with the length of the chain, while the costs of HC-Store and EC-Chain are several orders of magnitude lower. The cost of FR nodes increases linearly, while the cost of HC-Store nodes remains at a significantly lower and more stable level. Furthermore, Figure 8 and Figure 9 This demonstrates scalability as the network grows. The single-node costs of FR and HC-Store are stable, but the cost of EC-Chain fluctuates due to its group rebalancing mechanism. Crucially, across all scenarios, HC-Store consistently maintains lower and more stable storage overhead than the EC-Chain baseline model. These results strongly support the claim that HC-Store's total complexity is [missing information - likely OCR error]. The complexity of a single node is Its propositions highlight its outstanding efficiency.

[0225] C. Data Availability Verification

[0226] These experiments, under different failure and attack scenarios, verified the data recovery probability model of Theorem 2. Figure 10The query success rate was evaluated as the probability of node unavailability (due to benign failures or malicious denial-of-service attacks) increased. Results showed that HC-Store consistently maintained a higher success rate than the baseline EC model, especially with higher redundancy (…). In the case of ), Figure 11 An assessment was conducted on the malicious node deleting a portion of its stored data (the proportion was...). The impact of this is evident. In this scenario, compared to the EC-Chain configuration, HC-Store also demonstrates more stable and superior performance, proving the robustness of its integrated design.

[0227] D. Performance Verification

[0228] These experiments validated the performance model of Theorem 3. Figure 12 The on-time recovery probability under a given query deadline was evaluated. The results show that even with 30% malicious nodes, a success rate of over 98% can be achieved within a 200ms deadline, demonstrating the effectiveness of the parallel retrieval strategy. Figure 13 The total bandwidth overhead required to repair a single block for different k values ​​is shown. The box plot demonstrates that the repair cost is strictly limited and predictable, as confirmed by experiments. It is independent of network size and only expands with k, as predicted by Theorem 3(b).

[0229] E. Security Verification

[0230] These experiments verified the safety guarantee of Theorem 4. Figure 14 By comparing the profits of an honest node with two Sybil attack strategies, the system's resistance to Sybil attacks is visually demonstrated: one is the "compliant Sybil," which honestly fulfills its duties to verify the profitability of distributed equity; the other is the "detaining Sybil," which evades its duties in an attempt to manipulate the punishment system. The results clearly show that the detaining strategy leads to catastrophic financial losses, while the compliant strategy offers no profit advantage, thus proving that Sybil attacks are economically irrational. Figure 15 The detection model of ePoSt was validated. The figure shows the probability of detecting malicious data deletion as a function of the number of challenges (for different data deletion ratios α). As predicted by Theorem 4(a), the detection probability rapidly approaches 100%, confirming the effectiveness of the auditing mechanism.

[0231] F. Economic Model Validation

[0232] These simulations validated the long-term sustainability model of Theorem 5(b). Figure 16This shows the break-even price of user queries required to maintain the system's solvency, given a fixed number of queries and node rewards. Figure 17 This shows the break-even query volume required given a fixed user price. Both metrics evolve as the chain grows, experimentally confirming the theoretical conclusion that total user fees must eventually scale in sync with the chain's history. To ensure long-term sustainability.

[0233] The HC-Store of this invention is a novel layered storage architecture designed to address the key challenge of "storage bloat" in permissionless blockchains, thereby enhancing their long-term scalability and decentralization. Theoretical analysis and experimental evaluation show that HC-Store's heterogeneous node two-layer design significantly reduces storage overhead; simultaneously, its unified accountability framework ensures data integrity and availability without relying on altruism, exhibiting superior performance and resilience compared to traditional full replication and simple erasure coding models. HC-Store's total network storage cost is independent of network size, scaling only with its historical length; the theoretical models regarding efficiency, security, and performance have also been accurately verified by extensive experimental results.

[0234] On the other hand, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the hierarchical storage method for permissionless blockchains provided by the methods described above.

[0235] In another aspect, the present invention also provides a computer program product, the computer program product including a computer program that can be stored on a non-transitory computer-readable storage medium, and when the computer program is executed by a processor, the computer is able to execute the hierarchical storage method for permissionless blockchains provided by the above methods.

[0236] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0237] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0238] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A hierarchical storage system for permissionless blockchains, characterized in that, include: The hot node layer, consisting of multiple hot nodes, is used to store the hot data of the blockchain; The cold node layer, consisting of multiple cold nodes, is used to store both cold and hot data in the blockchain. On-chain smart contracts include: The dynamic data management module is used to periodically divide cold data into blocks and map them to the cold nodes for long-term storage using a consistent hashing algorithm; wherein, the cold data blocks are obtained by encoding the cold data; The storage proof verification module is used to verify the continuous storage status of the cold node for the cold data block and obtain the verification result; The incentive settlement module is used to automatically implement economic incentives for the cold nodes based on the verification results, and to feed back the results of the economic incentives to the dynamic data management module. The hot data includes blocks from the most recent W periods and current state data; the cold data includes all blocks from the previous W periods, where each period consists of a fixed number of consecutive blocks; each cold node is represented by a tuple consisting of a unique identifier and locked stake.

2. The hierarchical storage system for permissionless blockchains according to claim 1, characterized in that, The dynamic data management module includes: Member management unit, used to determine the set of active cold nodes in each cycle; A data allocation unit is used to allocate the cold data blocks to the cold nodes in the cold node set based on a weighted virtual hash ring. The data handover unit is used to update the ownership of the cold data block when the cold node joins or leaves.

3. A hierarchical storage method for permissionless blockchains, characterized in that, The hierarchical storage system for permissionless blockchains as described in any one of claims 1 to 2 includes: In response to each periodic boundary block, the dynamic data management module divides cold data into blocks and maps them to the cold nodes for long-term storage using a consistent hashing algorithm. The storage proof verification module verifies the continuous storage status of the cold node for the cold data block and obtains the verification result. Based on the verification results, the incentive settlement module automatically applies economic incentives to the cold nodes and feeds back the results of the economic incentives to the dynamic data management module.

4. The hierarchical storage method for permissionless blockchains according to claim 3, characterized in that, The step of mapping cold data blocks to cold nodes for long-term storage using a consistent hashing algorithm specifically includes: Each of the aforementioned cold data blocks is mapped to a logical slot in the virtual hash ring via a hash function; Based on the locked stake of the cold node, a virtual node is allocated on the virtual hash ring for each cold node, wherein the number of virtual nodes is proportional to the locked stake; The virtual nodes are placed at equal intervals on the virtual hash ring; The logical slot is assigned to the first virtual node encountered in a clockwise direction on the virtual hash ring, and the cold node that owns the virtual node stores the cold data block.

5. The hierarchical storage method for permissionless blockchain according to claim 3, characterized in that, The cold data blocks are obtained by encoding the cold data, including: The cold data is encoded using Reed-Solomon RS(k,m) erasure coding. A standard set consisting of the aforementioned cold data blocks; where k is the number of original data blocks and m is the number of redundant blocks.

6. The hierarchical storage method for permissionless blockchain according to claim 3, characterized in that, Based on the verification results, economic incentives are automatically applied to the cold nodes, and the results of the economic incentives are fed back to the dynamic data management module, specifically including: Based on the verification results, the proven storage share of the cold node is determined; Based on the locked rights of the cold node, determine the virtual node weight share of the cold node; Based on the weight share of the virtual nodes, basic rewards are distributed to the cold nodes that participate honestly. Based on the proven storage share, service fee rewards are distributed to the cold nodes that participate honestly. Based on the duration of failure to submit valid storage proof, a progressive penalty is imposed on the cold node, and feedback is sent to the dynamic data management module. When the duration exceeds the maximum preset tolerance threshold, the cold node is forcibly removed.

7. The hierarchical storage method for permissionless blockchains according to claim 3, characterized in that, The verification of the continuous storage status of the cold node for the cold data block specifically includes: At the start of the cycle, a globally unique and unpredictable initial challenge is generated based on a verifiable random function; Based on the initial challenge, each cold node sequentially generates a storage proof by sampling the cold data blocks in multiple rounds within a period, and forms a time-locked evidence chain based on the storage proof through a verifiable delay function; At the end of the cycle, a converged proof is formed based on the aforementioned chain of evidence; The on-chain smart contract performs a single verification of the aggregated proof.

8. The hierarchical storage method for permissionless blockchain according to claim 5, characterized in that, The method for retrieving the cold data after it has been stored in the cold node further includes: Obtain the set of active cold nodes; Based on the set of cold nodes, multiple cold nodes are determined to store the cold data blocks to be retrieved; Send data requests to multiple cold nodes in parallel; When each cold data block arrives, verify the integrity of the cold data block; iterate until the number of cold data blocks reaches the original number of data blocks; Based on all the aforementioned cold data blocks, the cold data is reconstructed through decoding.

9. The hierarchical storage method for permissionless blockchain according to claim 7, characterized in that, The storage proof is a storage time proof.

10. The hierarchical storage method for permissionless blockchain according to claim 4, characterized in that, The step of placing the virtual nodes at equal intervals on the virtual hash ring further includes: The location of the first virtual node is determined by the unique identifier of the cold node that owns the virtual node.

Images