File storage is a big matter in our world. People need to buy expensive hard drives to store data, or they have to buy online storage like Google Cloud Storage, Dropbox, or Amazon’s AWS.
Buying hard drives have some issues. Firstly, hard drives are expensive. Secondly, there is a great possibility of data loss and hard drive being burnt out that will lead to data loss.
In the other option, buying online and cloud storage may lead to other issues. These big storage providers use your data in many various ways such as analysing and selling them to other companies.
Decentralised storage is another way that may come to mind. In this option, storing files online will cost a bit more than centralised cloud storage but it is much safer, because no one has access to all of your data to analyse and sell it.
Crust Network is a way to make decentralised storage come to life and it can be efficient for users to use it.
Crust provides a decentralized storage network of Web3 ecosystem. It supports multiple storage layer protocols such as IPFS, and exposes storage interfaces to application layer. Crust's technical stack is also capable of supporting a decentralized computing layer. It is designed to build a decentralized cloud ecosystem that values data privacy and ownership.
Crust can be used in a wide range of scenarios, including:
Serverless DApp/website hosting
Developers are able to deploy their own Dapps/websites based on Crust/IPFS Network. With this, developers are able to only focus on building Dapps/websites, no need for server management. Moreover, Crust Network incentivizes multiple nodes to store Dapp/website files and provide storage services, thus ensuring the high availability of Dapps/websites. [Crust Apps] is an example, it is hosted on Crust Network.
General off-chain data storage
Some blockchain-based decentralized applications, such as NFT and decentralized social platforms, involve a large amount of user data, far beyond the on-chain storage capacity. Based on Crust/IPFS Network, developers can provide general off-chain data storage services for these applications.
Accelerated content delivery
Developers are able to build P2P content delivery network (CDN) services with Crust/IPFS Network. Crust Network incentivizes the nodes to spontaneously distribute CDN data and play the role of CDN edge nodes to accelerate the data transmission for CDN users.
Secured cloud storage
Developers are able to build Cloud storage applications with Crust/IPFS Network. Without maintaining storage clusters, developers only need to integrate the SDK of Crust/IPFS into applications to provide users with stable storage services.
Crust works in 3 layers as follows.
MPoW (Meaningful Proof of Work) - The low-trust/zero-trust storage proof layer to inspect and prove storage work of miners.
GPoS (Guaranteed Proof of Stake) - The PoS-derived consensus layer that requires nodes to provide storage proof to get staking quota.
DSM (Decentralized Storage Market) - It enables users to place storage orders on the chain, to get their data stored and distributed on Crust Network.
A) TEE (Trusted Execution Environment)
Trusted computing refers to applying trusted computing platform on HSM (Hardware Security Module) support in computing and communication systems to improve system security. With the deepening of trusted computing studies, the public attention has gradually shifted from the traditional hardware chip security model to TEE. TEE is a concept proposed by Global Platform. Currently, TEE has a variety of implementation solutions, among which Intel chip-based SGX and ARM open source framework-based TrustZone are the two most widely recognised and applied in TEE technical implementation.
TEE features a secure combination of multiple computer-related technologies. The following 5 technical concepts are the core specifications of TEE:
B) MPoW Mechanism
There is no such centralised institute like a bank in the blockchain system, instead, it is the consensus mechanism that guarantees the consistency and correctness of each transaction on all accounting nodes in information transmitting and value transferring. The consensus mechanism of the blockchain enables all nodes of the whole network to work in large-scale collaboration without relying on any centralised organisation. Some current mainstream blockchain consensus mechanisms, such as PoW and PoC, often need to rely on workloads calculated through specific computing or storing procedures that are generally considered meaningless. Crust, however, by combining TEE technologies on the basis of distributed storage and verification scenario, has proposed the original MPoW (Meaningful Proof of Work) mechanism. Compared with those current meaningless space-based solutions, MPoW can be used to quantify a variety of meaningful data storage and computing process in a secure, fair and efficient way.
MPoW is a low-trust storage proof layer. The overall function of MPoW is to periodically generate trustable storage work reports and upload to the chain. A storage work report includes incremental information of data storage work on a particular node.
The MPoW mechanism is mainly responsible for node workload calculation and environment verification. We will explain these two functions and related processes from a storage scenario:
Workload calculation: A node receives distributed data and stores it to the hard disk. When users’ data are stored, a regular spot check program is performed in the local TEE, which aims to verify the Merkle Hash and check whether the storage space declared by the node is used to properly save users’ file or not.
Environment verification: A checking program is run in the node TEE to remotely attest the logic of other nodes’ TEE environment information and trusted execution code version information.
It can be seen that the integrity check for the data, the verification and calculation of the storage, the examination of the node environment and the identity verification of the nodes are all under the sound protection of the TEE.
MPoW features the following advantages:
Transparency: The storage mechanism is both open and transparent.
Fairness: The computing for the workload and reward of storage node both are under TEE protection, and nodes are freed the angst about their workload being unrequited. Also, no extra rewards will be gained by cheating.
Efficiency: Extra redundancy and storage of meaningless data both are kept away from the proof of storage. Both computing resources and storage resources are utilized efficiently.
Evolvability: The TEE supports complete computing and is potentially evolving. This means that Crust blockchain ecosystem can be based on MPoW to perform more powerful functions, ensuring the evolutionary feasibility from storage consensus to computing consensus.
Based on the Trusted Execution Environment(TEE) technology, nodes in Crust network can perform non-interactive storage proof. As shown in the figure below, TEE will periodically check the storage status of nodes and generate a workload report that can be verified by all nodes.
GPoS is a PoS-derived consensus protocol. Like NPoS, a user can vote to specific nodes (and we define the voting behavior as a guarantee process). The key difference is that each node has its own staking quota, decided by its proven storage resource and work. Therefore, GPoS encourages nodes to provide more storage resources to take more work, and improve the security and robustness of the network.
Multiple participants are seen in the entire Crust system, each having varied needs. According to the way of participation, they can be divided into: verifiers, candidates, guarantors, and users (to be specific, users of storage and computing resources).
Verifiers: Verifiers are nodes that are responsible for packaging and generating blocks in Crust network while also maintaining the entire blockchain network. According to the GPoS (Guaranteed Proof of Stake) consensus of Crust network, verifier nodes need to hold some storage resources as a guarantee, and they can stake CRU tokens while staying online. Therefore, verifier nodes are also where storage resources are provided. Besides, verifier nodes participating in the network can obtain the rewards separately given to block package and the reward share of each blockchain cycle, also bearing the risk of assets being confiscated. Verifiers can also sell storage resources to gain income in the storage transaction market.
Candidates: Candidates are nodes that participated in the competition of verifiers, but were not in the end qualified for verification in Crust network. Similar to verifier nodes, candidate nodes also need to have storage resources as a guarantee, and can stake a certain number of CRU tokens while staying online. The difference is that candidate nodes do not participate in the block generation and thus cannot obtain the reward separately given to nodes that generate blocks. But candidate nodes can get the reward share of each blockchain cycle, and they can also sell storage resources to gain income in the storage transaction market. It should be noted that candidates and verifiers are not fixed. They may change every cycle, which is mainly determined by the number of tokens staked by nodes at the end of each cycle.
Guarantors: Guarantors are accounts that provide guarantee for anyone or a few nodes in Crust network. Any account with CRU tokens can become a guarantor, and its CRU can be used as an encumbered asset. Guarantors can also obtain guarantee income by providing guarantees for nodes.
Users: Users are consumers who use Crust network resources, mainly those using storage and computing resources. Users can use CRU tokens or other token assets available in Crust network to purchase resource services.
Crust chain has adopted a GPoS (Guaranteed Proof of Stake) consensus mechanism, which is also known as PoS consensus with storage resources as guarantees. Similar to existing PoS projects, nodes need to compete for the position of verifiers by staking CRU tokens, while the difference is that nodes additionally need to provide storage resources in order to obtain corresponding guarantee limit which makes staking CRU itself possible in the first place. Through the MPoW mechanism and node storage capacity monitoring mechanism, it is the case that the more storage resources a node contributes, the higher the limit will correspondingly be.
GPoS leverages the storage work reports generated by the MPoW storage proof layer to calculate the storage resource related parameters.
GPoS also encourages users to stake their tokens to high quality nodes via guarantee operation to obtain staking income.
GPoS related concepts and processes are described in the following diagram.
Crust DSM (Decentralised Storage Market) aims to provide high-quality storage services for applications and platforms based on the Crust network. Storage services mainly involve a storage order mechanism and are trieval mechanism.
Pricing Mechanism: In the Crust network, users sign storage orders regarding the entire net work rather than specifically centering on a single node. In this mode, when a user storage order is generated, a corresponding price will be calculated by the network in relation to the current storage supply-demand situation.
Storage Order Mechanism: DSM provides users with access to order storage, and users can store their files in the Crust network on a long-term basis with some payment.
Crust storage order mechanism is pool-based. With this mechanism, a user generates an order containing information of storage demands and a short description of the files to be stored to Crust network. One part of the fee paid by the user will go to the reward pool of the entire network where it will be distributed to nodes that provided CRU token staking; the other part will flow to the reward pool, where the user’s files were stored, and be paid to those who provided storage proofs for the files.
Nodes in the Crust network can obtain the files through IPFS and save them locally. Then, nodes can declare file storage shortly after the sealing, verification and proof by the local MPoW. All nodes that provided file storage proofs will enter the reward list in order, and those top-ranking ones will obtain rewards from the reward pool.
The efficiency of a node obtaining files and MPoW providing proofs will have an effect on the node’s ability of order taking. Responding to this, Crust network adopts a credit mechanism similar to BitSwap to enable highly capable nodes to obtain user order files more efficiently.
Storage services of Crust network are mainly adapted to such technologies as Inter Planetary File System (IPFS) and Distributed Hash Table (DHT), enabling basic data integrity, content addressing, tamper resistance and deduplication. The difference is that by the strength of MPoW, storage capacity computing and verification can be performed in the local TEE, which greatly increases the efficiency and reliability of workload computing.
Apart from those basic storage features, Crust Network took a step further in user privacy protection. The TEE-based Crust network can support the establishment of encrypted channels and the sealing of data between node Enclaves. Users can decide their private data to be transmitted through encrypted channels and to be stored on a TEE sealed manner. User data encrypted in this way will not be obtained by anyone (even storage nodes themselves) other than the only user.
Retrieval Services: There are two types of retrieval demands in the Crust network: those from users and those from nodes.
The former reflects users’ needs for data use, which is also a manifestation of the value of storage applications on the Crust network.
Similarly, nodes also have retrieval demands for files. For one thing, the competition for rewards from new orders is based on retrieval; For another, storing meaningful files can extend the Stake Limit of a node.
Crust network nodes draw on the credit game mechanism similar to BitSwap, that is, to provide more retrieval services to those nodes that have ever offered services. Consequently, the result of the game is that nodes which provided retrieval services will be more likely to retrieve data from other nodes.
It is also in such a game mechanism that an incentive cycle has been formed between the storage and the retrieval of meaningful files. The income of a node is contingent on how efficiently the node can retrieve files, which is further determined by the node’s response to previous retrieval requests. For nodes in the Crust network, they will store as much user data as possible and increase their data download speed, a way user experience has been greatly improved.
DSM contains 3 major mechanisms:
Pricing Mechanism. Storage order fee is calculated by the network according to file size, storage period and overall storage requirement and supply.
Storage Order Mechanism. Storage Order works in a pooled way. Once a user places a storage order: a) Certain amount of order fee will be added to the specific file payment pool, and the rest will be put to the overall network staking pool. b) Any node that submits corresponding storage proof will be added to the payout queue of the file payment pool, until the payout queue reaches its limitation. Nodes in the queue will be able to claim the income later. c) Any node that submits corresponding storage proof will get a staking quota increment (by several times depending on the file copy amount all over the network).
Retrieval Mechanism. As a data storage system, besides the storage market, DSM also takes the responsibility of incentivizing the nodes to provide retrieval service. Therefore, DSM includes another layer of incentives to encourage nodes to support data retrieval: a) DSM enhances the IPFS credit system, which is a local node ranking system without consensus, as part of the BitSwap strategy. In a Crust-backed BitSwap process, a node is able to query the on-chain storage proof info to get what blocks are stored on which nodes; therefore, if it failed to retrieve blocks from target node, the local ranking for target node will be lowered down. b) According to the credit system, nodes providing better retrieval service will get higher ranking and have a better chance of getting more files quickly, thus getting more storage payments and staking income. On the other hand, nodes that provide lower quality of retrieval service will unlikely to benefit from the system.
One important point to notice about Crust Network is that even though its mainnet is up and running, the only way to connect to the project and store your files to other people’s storage is not through the Crust Network’s mainnet. It is possible for you to use Crust Wallet, EVM-based wallets like Ethereum, Polygon, BSC, HECO, etc. Other web 3.0 wallets like Near, Elrond, MetaX, Solana, Flow, and Polkadot Wallet.
From technological view, Crust network is a good solution to the decentralised storage problem, but I don’t understand why someone should build this platform when FileCoin is up and live and working well almost exactly the same way.
The Crust Network is coded using Substrate, Polkadot’s Programming language. Crust is also sitting on one of the Polkadot’s parachains. Crust has little number of commits and it is updated very rarely. Number of contributors are a few and this is a bad news for the future of the protocol.
The only audit that I found for this platform is from a not well-known website named Certified which haven’t fully researched on the protocol and gave a D grade (bad grade) to the protocol.
Anyways, I personally don’t give a good grade to this part.
As my researches told me, Crust Network is a product of Decentralized Cloud Foundation. This foundation is mainly responsible for crust network being launched and working. In this team I found some team members worth watching and studying.
Leo Wang is the permanent director of technology at Decentralized Cloud Foundation. He is the co-founder at Clouderium Technologies and he has some CTO and Technical Lead in his resume. He got his BS in Computer Science from Harbin Institute of Technologies. He is obviously a key member and can be trusted.
Luke Fan the head of research at Decentralized Cloud Foundation. He works as CTO at Clouderium Technologies. He has a background of software Engineer at Microsoft and got his Computer Science degree from Zhejiang University.
Other than all of the web 3.0 wallets that Crust Network supports, there are other partners that Crust Network works with. From these companies and projects we can name Apron Network, Ares Protocol, Automata Network, Cere Network, OpenSquare Network, Tidal Finance, Khala Network, Phala Network, Manta Network, Crab Network, etc.
Partners are not that bad but they are not very reliable to grow Crust Network.
There are some investors backing Crust Network. Web3 foundation grants program, Substrate Builders Program, DFG, NGC ventures, Hash Capital, SevenX Ventures, IOSG Ventures, Bitcoin.com, BR, Chain Capital, Bitscale Capital, AU21 Capital, SNZ, 3Commas, Consensus Labs, Origin Capital are the list of investors for Crust Network.
The list of investors is long but almost none of them are reliable and big.
Looking at the roadmap we can see that they have very great plans for the future and we can give a good grade to the project for this part.
As a decentralized cloud system, Crust serves as a distributed network that is public and open to everyone to participate freely. This system was incubated by the Decentralized Cloud Foundation, and it will eventually be self-governed by the community after the growth period. In this system, whether it is a cloud service provider or demander, or various interested parties who maintain the system, they can freely enter and exit under the premise of following the system agreement. Crust's economic model can maintains the interests of all parties and guarantee the development of the entire Crust system.
There are multiple parties in the entire Crust system, and they have different needs. According to the way each role participates, we divide them into validators, candidates, guarantors, and users. The users mentioned in this article mainly refer to users involved in storage and computing resources. At Layer 2 there will be other different user roles, such as a token market maker.
The native token CRU in the Crust network is a utilitytoken representing the value of the entire network, similar to ETH in the Ethereum or DOT in the Polkadot.
Token Functions
In the Crust network, the CRU token mainly has the following functions:
Generation and Burning
There are two ways to generate CRU tokens: one is generated at one time when the main network is initiated; the other is generated as blocks are produced.
The number of genesis blocks as Crust initiated main network: 20,000,000 CRU
Mainly used in the following aspects:
The tokens generated by the block have rewards every cycle, and the points can also be redeemed for rewards. The rewards are mainly for participating nodes in the network to maintain the security of the network protocol. And encourage participation in the network at an early stage. The total distribution method is as follows:
If the service quality of the node is unstable or node is found with malicious behavior, it will face the confiscation of the staked CRU. Some of the confiscated tokens will be directly burned, and some will be placed in the Treasury account as a reserve. Some of the fees incurred by transactions in the network will be directly burned, and the rest will be allocated to the nodes that generate blocks.
For a brief conclusion and my opinion on the Crust Network please see TL;DR section.