An article detailing the head-to-head battle between dark horse Swarm and frontrunner Filecoin

Swarm is a distributed storage project built by the official of Ether. This article explains the essential difference with Filecoin through economic model, technical analysis, incentive mechanism, functional value and other aspects.

With the opening of the bzz airdrop program, more and more people are learning about Swarm from various sources and communities and are interested in the project. According to rumors, Swarm is considered to be one of the three pillars defining Web 3.0 components, along with Ether and Whisper. What is Swarm? What are the advantages? How does it differ from Filecoin in terms of distributed storage? Next, Ice will take you to find out.

Who is Swarm?

First of all, let’s understand what Swarm is: Swarm, as a member of the original ecology of Ethernet, is an official part of the Ethernet project, which is mainly led and developed by the Ethernet Foundation and uses different protocols and technologies in the Ethernet blockchain. It allows cloud pooling of storage, bandwidth and arithmetic resources to support applications based on the Ethernet network.

Secondly, Swarm provides a more friendly way of development, and Ethernet developers can accomplish the task of decentralized storage of data through Swarm, instead of relying on external ecology directly. Meanwhile Swarm decentralized content storage and distribution service, which can be considered as a CDN and distributed mainly through computers, can run Swarm nodes and connect to the Swarm network as if they were Ethernet nodes.

Once again, the Swarm team is trying to create a no-downtime, zero-failure and censorship-proof peer-to-peer storage and service solution. Creating a financially incentivized system within Swarm will facilitate the payment and transfer of the value of the resources exchanged. This is similar to BitTorrent, but can also be seen as IPFS, using BZZ as a reward incentive: data files are broken into chunks and assigned to participating storage providers to store them, and the providers who provide storage and retrieval services for the chunks receive BZZ as a reward from the nodes that need the storage and retrieval services.

Finally, Ethernet’s smart contracts implement distributed logic; Ethernet’s Swarm implements distributed storage; Ethernet’s Whisper implements distributed messaging to interoperate between smart contracts, thus enabling more complex DApps. Swarm will also become the hard drive of the world’s computers, enabling the Internet to be decentralized again.

Swarm’s biggest advantage over Filecoin, which is also in the distributed storage track, is that it has ready-made smart contract ecological resources and ready-made effective hot data storage, which can quickly realize the value of its distributed storage. At present, the ecology of Filecoin is still in the newly launched stage, and most of the data stored in Filecoin is garbage data, and the real effective ecological data still needs some time to develop to gradually present. For other aspects of the comparison, Ice will elaborate from the following aspects.

※Swarm and IPFS team’s genius duel

  • Swarm -V God stand endorsement

Back in early 2015, the concept of Swarm to implement distributed data storage was proposed. Swarm has become a star attraction and a popular project, thanks to the fame of Ether and Vitalik, who created Swarm’s protocol tags bzz and shh.

Previously, Swarm was spun off from the Ether Foundation and became an independent organization with a core team of 32 people currently divided into 8 groups: Leet Squad, Bee Team, Bee-JS Team, Comms, HR, DevOps, Ops, and Knowledge Management.

An article detailing the head-to-head battle between dark horse Swarm and frontrunner Filecoin
  • Filecoin – Protocol Labs

Founded in May 2014 by Juan Benet, the inventor of IPFS and Filecoin, Protocol Labs joined the top US incubator Y-Combinator in 2014, and in January 2015, Protocol Labs released IPFS to the world. IPFS has become one of the most popular technologies in the blockchain industry, and the thousands of developers at Protocol Labs have been called the “future of the WEB”. Protocol Labs has also created libp2p, IPLD, multiformats, Orbit and other projects.

Swarm and Filecoin’s vision go head to head

  • Swarm – Rethinking the way the network is built

Swarm’s vision focuses on “wanting to build a decentralized storage and communication network on the blockchain”. It was birthed long before the main ethereum network went live and was proposed as one of the pillars to support Vitalik and his team’s vision of the ultimate decentralized network.

Today’s Internet is gradually becoming dominated by large centralized companies and organizations such as Google, Facebook, Amazon, etc., resulting in a huge threat to online data privacy. With the huge amount of user data, algorithms are used to manipulate all kinds of information received by users on the Internet and monitor their daily network. The birth of Bitcoin and Ether opened a window for decentralized Web 3.0, but they both focused more on systematic data transactions and less on the storage of non-systematic data. Swarm was born to solve this problem by rethinking the way the network is constituted and supplementing it with a foundation of economic incentives for pass-throughs.

  • Filecoin – a key building block for Web 3.0

IPFS is benchmarked against the HTTP protocol of the Web 2 era, while Filecoin, as the “official incentive layer” of IPFS, is a distributed storage network built on the basis of combining IPFS and blockchain to secure data information, break data silos, and allow data information to flow freely. Filecoin brings “trust” to the existing Internet development and is positioned as a storage infrastructure in the Web 3.0 era. The reliability of file storage.

Filecoin is the embodiment of the value of the IPFS network: all value services on the IPFS network are transacted through Filecoin, and storage service providers are rewarded with block packing through the business of providing storage space, while users pay FIL to storage service providers through their storage and retrieval needs. filecoin serves as the incentive layer of IPFS, and anyone who only wants to store files can join Filecoin, and the price of file storage is not controlled by anyone, users only need to find their own satisfactory file storage price. At the same time, Filecoin’s network mechanism, any storage service provider can publish their own file storage prices, and users can also find by looking for their own satisfactory file storage prices.

The birth of Filecoin in the Web 3.0 era is a revolutionary technology that builds a reliable Web 3.0 infrastructure. It shifted the previous centralized model to a decentralized development model. The resulting benefits are that it removes centralized nodes by decentralizing them, greatly reduces the risk of being censored, and increases the possibility of being attacked. the emergence of Filecoin allows each individual to truly control and use their own data, establishing their own access to their own data.

Swarm and Filecoin’s Innovation Competition

  • Swarm – A new way of storage

Swarm aims to build an autonomous digital society where data is stored and communicated without the need for permission from anyone other than the individual, resulting in a truly free and open web environment. World Wide Web (WWW).

In addition, Swarm creates a unique incentive system for data storage and distribution, ensuring that the costs and rewards involved in storing and distributing data are reasonable through swap agreements and incentive systems such as wear and swindle, Swarm effectively solves the pain point of BitTorrent’s lack of economic incentives. And BZZ, as a functional pass for Swarm, is perfectly applied to its incentive system.

The future growth of Swarm lies in its holding (i.e. production), liquidity and self-growth, and the nodes eligible for BZZ passes are those that have been proven to be in line with what officials call trusted “queen bee nodes” (“queen bee nodes – qbzz nodes “that is, the official randomly designated node users, currently 35) have exchanged bandwidth nodes, through the official information we can understand that proven nodes (valid nodes) with the official trusted nodes (queen bee nodes) to produce proof of bandwidth data exchange in order to obtain proof of ticket stubs (qbzz checks) for airdrop BZZ, qbzz checks in order to exchange BZZ. Scrap tickets really don’t make much sense for investors compared to the number of non-qbzz checks that are obtained. Currently Swarm bee nodes are located all over the world, and no one can be completely sure of the exact location of the node behind the qbzz bee during the airdrop phase, but the greater the contribution to the network, the greater the chance of getting BZZ.

An article detailing the head-to-head battle between dark horse Swarm and frontrunner Filecoin
  • Filecoin – A New Proof Mechanism

Filecoin has developed an innovative hybrid consensus mechanism – Proof of Replication (PoRep) + Proof of Space and Time (PoSt) + Expected Consensus.

Filecoin requires proof of replication (PoRep) to confirm that the data has been sealed and stored properly, and that proof of replication can effectively prevent witch attacks, outsourcing attacks, and generational attacks.

Proof-of-time for Filecoin = windowPoSt + winningPoSt. Proof-of-time (PoSt) is essentially a type of PoS, but it can reduce the waste of resources to a certain extent, save computing resources and energy, and use the size of the data stored on the hard disk as arithmetic power, instead of performing huge amounts of hash calculations.

① windowPoSt means that a node needs to submit a windowPoSt message every 24 hours in the valid sectors of the day to prove that the data is correctly and properly stored in the sector.

② winningPoSt then refers to the PoSt message that needs to be completed to prove that it has the ability to get the block right when it gets the block right.

Filecoin’s Expected Consensus (EC): The goal of EC is mainly to make the node’s block-out entitlement proportional to the corresponding storage contribution.

The difference between Swarm and Filecoin’s economic model

  • Swarm – Build a perfect incentive mechanism to promote user motivation

BZZ is Swarm’s functional (UTILITY) Token, which is also used for bandwidth and storage rewards.

BZZ initial supply: 62.5M.

Token allocation.

Fundraising : 50%

Team : 20%

Foundation : 7%

Infrastructure Grants : 10%

DApp Grants : 10%

Donations : 3%

Swarm will be running a test network from February 14, 2021 until the main network goes live (expected in 2Q21). 1 million BZZ passes will be given out on the website to reward early adopters and stress test the network. Participating nodes will receive checks periodically, and by the end of the test, exactly 35 “queen bee nodes” (qBZZ nodes) will eventually be identified in the network, and the checks (i.e., qBZZ checks) earned by interacting with these 35 qBZZ nodes will be valuable, but no one (not even team members) will know about them until the end of the test. even team members) will know which nodes these are. The official airdrop gives away exactly 1 million BZZ passes. At the end of the airdrop, officials will check how many qBZZ checks are available across the network and determine how many nodes are entitled to BZZ, as well as the qBZZ/BZZ ratio, which is uncertain for now until the airdrop ends.

Swarm’s incentives are divided into the following three categories.

  1. Bandwidth incentive

Bandwidth is used as a hard requirement in the data exchange of Queen Bee nodes. swarm uses bandwidth incentives to build trust relationships to exchange bandwidth, facilitate data uploads and downloads. bee nodes maintain fractions of consumed and received bandwidth as they communicate with each other to facilitate private and secure uploads of content.

  1. Discovery Incentives

Swarm’s discovery incentive is indirect, and it is in the interest of each node to have a well-connected Swarm network. For example, when new Bee nodes appear, they will be connected to existing Bee nodes, and the number of connections on a Bee node is expressed as a “connection level”. When the node connects to another node, it also communicates its connection level, and then another (older) node can suggest new connections for it to increase the connection level of the new node.

  1. Storage incentive

Swarm’s storage space is limited by the sum of all individual nodes’ storage contributions to the network, and storage incentives are set to ensure fair compensation for those who store. Allocating the right to write to Swarm in the best way possible addresses the issue of increased costs due to the need for the postage stamp system by uploading costs to the Swarm network. swarm storage nodes can decide for themselves what content to keep and what content to ignore, allocating storage space in a way that retains the most valuable of those blocks, and using money paid through the postage system, the project compensates nodes that can prove they are storing data for the node.

Filecoin – Sustaining supply and demand through block rewards and pledging mechanisms

Filecoin has one of the more complex economic models of any blockchain project. Since the project’s inception, its economic model has undergone thousands of modifications and its economic parameters are constantly changing. After going online from the main web and experiencing problems such as economic model changes, pledge parameters, Gas fees, and main web outages, it finally stabilized and has been continuously optimized and improved through various proposals.

Token Allocation.

FIL total issue volume is 2 billion pieces

Infrastructure Grants: 70%

Protocol Labs: 15%

Fundraising: 10%

Foundation: 5%

Filecoin’s Block Rewards

Filecoin storage providers are rewarded with blocks in the storage marketplace based on their storage capacity. Storage capacity is measured by how many units of active storage space a storage provider provides in the Filecoin network and how long it can store them, and the storage capacity of a storage provider in the Filecoin network is equivalent to the computing power in the Bitcoin network. The block bonus a storage provider receives depends on the percentage of its storage capacity to the network-wide storage capacity. For example, if a storage provider provides 1 PB of storage capacity in a network of 100 PB, it will receive one percent of the block bonus for that amount of time.

The main difference between Filecoin and other projects is the nature of the service provided: storage is a long-term service, and Filecoin requires storage providers to keep the network stable over time, unlike the Bitcoin network, where storage providers have free access. Therefore, Filecoin uses the Network Baseline reward model (Network Baseline s) rather than the time-exponential decay model. The Network Baseline model has two mechanisms.

First, block rewards decay exponentially in time over the long term.

Second, each block height corresponds to a network-wide storage capacity criterion. If the network-wide storage capacity does not meet the standard when the block reward is generated, some of the block rewards are deferred until the network-wide storage capacity reaches the standard and the reward is released.

The network benchmark reward model has two benefits for the Filecoin network.

First, it ties block rewards to storage market growth to avoid storage service providers from exiting the network after receiving large early rewards.

Second, it smooths out the exponential decay model to avoid massive network-wide storage capacity changes due to halving of rewards.

Filecoin’s pledge mechanism

Filecoin’s pledge mechanism appears to be a storage service provider penalty mechanism, on the one hand the future is to safeguard the data of Filecoin storage marketplace customers from being lost. On the other hand, Filecoin’s pledge system deepens the interest bond with storage service providers, extending the time cycle of Filecoin’s ecological construction and providing a clearer trend for future development.

In the storage agreement between each user and storage provider in the Filecoin network, the storage market provider will store the data provided by the customer in the network, called Sectors. Each sector added by the storage provider contains the contents of the stored files and the committed storage duration, ensuring that the customer is free to use their stored data for the agreed storage duration. When a storage provider adds a sector to the Filecoin network, the storage provider pledges a locked asset.

The locked asset consists of two components: the storage provider’s own FIL pass and a portion of the block reward. If the storage provider goes offline during the committed storage time, he loses a portion of the locked assets. If the storage service provider stops the storage behavior completely, he may lose all locked FIL. locked assets are unlocked after the storage service provider fulfills the storage agreement, and the storage service provider can recover all assets after a short lock-in period. Pledging part of the block reward has two benefits: the first is to avoid the early scenario where FILs are short of liquidity in the secondary market. The second is to reduce the cost of entry for storage service providers.

  • Swarm and Filecoin technical aspects of mutual cut
  • Swarm-Exit the circle by inheriting the advantages of Ether

Swarm’s deep integration with Ether provides Swarm with a large audience and user base, and enables the protocol to scale more quickly and effectively, providing infrastructure-level advantages not available with other protocols, whose development is also guided and inspired by Ether’s needs (most importantly, the need to host dapps, contracts, data, blocks, etc.).

On the one hand, Swarm provides a storage layer integrated with Ether’s compute layer and Whisper’s secure messaging layer.

On the other hand, Swarm is a local layer in the Ether stack that can be integrated with Ethereum’s browser, making it easier for regular Ether DApp users to interact with the protocol.

Swarm’s core storage component acts as an immutable content addressing chunkstore instead of a generic DHT (Distributed Hash Table), and users can use Swarm as a cloud host for uploading content to the Swarm network. In addition, Swarm is attached to the Ethernet P2P network and inherits its audited and widely acclaimed security properties, using Ethernet’s DevP2P (protocol multiplexing, framed message interleaving, encryption, authentication, handshake and protocol messaging API standards, peer-to-peer connection management support, node discovery) for message interleaving. This suggests that Swarm supports relatively novel solutions in terms of both motivation and storage or retrieval.

While Swarm, like IPFS, implements key-based routing based on XOR logarithmicdistance (applied to a shared address space of node IDs and content hashes), Swarm nevertheless uses a hybrid style of sending/forwarding (forwarding) Kademlia: instead of having the originator of the request rely on a larger pool of peers to perform iterative lookups and filtering, Swarm recursively outsources the successive lookup steps and uses only a smaller pool of active connections.

Filecoin – A unique proof system that stands out

Filecoin’s unique proofing system ensures that the marketplace works consistently and efficiently and that users can be confident that the storage providers on the network are storing data correctly and contractually. The integrity of data stored on the network.

On the Filecoin network, anyone in the world can provide storage space. But solving the challenge of trusting a decentralized storage network requires establishing a way to trust on the global network itself. filecoin’s proof system is achieved by establishing trust in the protocol layer of the Filecoin blockchain through storage proofs as the core mechanism for reaching consensus in the Filecoin blockchain.

On each public blockchain, “consensus” defines the agreement on the data entries that exist on the blockchain. To reach consensus, blockchains rely on consensus algorithms to generate new blocks and protect the blockchain from fraudulent data input. These algorithms determine precisely how nodes contribute to building and maintaining the blockchain. While other blockchains rely on consensus algorithms that require nodes to perform wasteful work (e.g., proof-of-work blockchains), Filecoin’s consensus mechanism engages nodes in useful work to provide and verify the integrity of stored data (various proofs of storage, i.e., replication proofs and spatio-temporal proofs).

At the heart of Filecoin’s consensus algorithm are two proof mechanisms that together make data storage publicly verifiable on the Filecoin blockchain: proofs of replication and spatio-temporal proofs.

Proof of replication (PoRep) starts with a process called “sealing”. The storage provider dedicates a portion of the available storage space, called a “sector”, to store the client’s data. After the sector is filled, it is sealed. Sealing is a set of operations that gradually converts a sector into a unique copy of the original data. This copy is associated with the public key of the Filecoin storage provider. To perform the final PoRep, the cryptographic hash of the copy (its CommR) is submitted to the public Filecoin blockchain.

The PoRep provides public proof that they are storing a unique encoding of the customer’s data at the time the proof is executed. Only those with all the original data can submit the correct CommR (on-chain commitment to a copy) to the Filecoin blockchain.

Since a PoRep does not verify that the storage will be continuous over time. This is where Proof-in-Time (PoSt) comes in. PoSt requires randomly selected storage providers to provide PoReps for randomly selected storage sectors they maintain. this is done through a process of issuing a cryptographic challenge to the storage provider, which can only be answered correctly by consulting the sealing department directly. The storage service provider must respond to this challenge within a strict timeframe. Each storage service provider must prove all of its storage on a daily basis and also randomly select storage service providers to prove storage in order to win blocks. The computational difficulty of sealing ensures that storage providers must maintain ready access to and integrity of the sealed area.

PoRep and PoSt together make up Filecoin’s unique proof system, a combination of proof of storage and proof of space.

The battle between Swarm and Filecoin incentives

  • Swarm-Built-in incentive system to meet transaction needs

Swarm has a unique built-in incentive system built around the three key components of Swap, Swear and Swindle, which enables various transaction needs of financial digital coins through user nodes through the Swarm network system. Swarm takes full advantage of smart contract functionality to handle deposit payments from registered nodes, allowing for coercive measures as a deterrent.

Swarm is a point-to-point content delivery system that redeems the final check by introducing checkbook smart contracts for off-chain delivery, while combining Swarm and Swap to solve the problem of possible duplication of checkbooks. Under the existing incentive system, Swarm can deliver content through nodes, and the Swarm CDN will be automatically scaled, and data with high attention and many views will be saved by each node, but the related fees need to be paid. As a key feature of content insurance, Swarm will implement efficient automatic collective auditing of rarely accessed off-chain content and final litigation on the blockchain, which may be removed over time if it is not very important.

In addition, using a paired accounting protocol and delayed micro off-chain payments, Swarm offers significant transaction cost savings while maintaining security. Since Swarm’s pairwiseaccounting for registration and final litigation, delayed payments, and collective audits are off-chain, it is possible to reduce the reliance on the blockchain to a certain extent.

Finally, Swarm’s concept of “Manifests” (common routing tables/key-value indexes with integrity protection) allows for.

  1. modeling hierarchical file systems on the cloud.
  2. serverless servers with routing tables and metadata schema systems (content types, encryption and insurance information, etc.).
  3. implementation of arbitrary DHTs inside Swarm, so it can support “side chaining” or db components of traditional webapps (e.g. mysql in LAMP stack, etc.).
  • Filecoin – Balancing supply and demand to facilitate the growth of effective storage capacity

Filecoin adds an incentive layer to IPFS and plans to incentivize the network through IPFS, aiming to use its altcoin blockchain and include proof of retrievability as part of it. filecoin storage marketplace operates in a dynamic balance between the various players. In the early stages the demand for storage did not match the supply perfectly. When the demand for Filecoin eco-storage was lower than the supply, storage service providers had two options: either to idle their devices and wait for the demand for transactions to arrive, or to store useless data for block rewards.

Filecoin uses two mechanisms to address the imbalance between supply and demand in the storage market.

Committed Capacity Committed Capacity

Committed capacity is a mechanism that allows storage providers to avoid idle devices. When a storage provider has idle equipment due to lack of storage demand, it can provide committed capacity sectors to the Filecoin network. The Filecoin network proves that the storage provider has valid storage capacity and can provide storage services to users, and the storage provider is rewarded with a block. Then, when a customer’s storage needs arise, the storage provider’s committed capacity sector can be immediately upgraded to a valid storage sector to store the customer’s data and receive a transaction fee. This mechanism can avoid the behavior of storage service providers to obtain block rewards by storing invalid data, and increase the effective storage capacity of the whole network.

Verified Datasets

The main purpose of the Verified Datasets mechanism is to prevent storage providers from gaining revenue through invalid data attacks, which deplete the storage capacity of the entire network. Data stored by users becomes valid data after being verified by the Filecoin network. Both storage providers and users are motivated to validate data, and storage providers can receive higher block rewards for storing validated data, while users will also increase the priority of storage providers in processing their own transactions by providing validated data.

In addition, Filecoin has no penalty storage, while Swarm has designed penalty storage. filecoin has no storage penalty mechanism, which leads to the proliferation of spam content, and even some Filecoin storage providers deliberately create spam content in order to increase revenue, while Swarm improves file availability and security by storing penalties. Although Swarm’s current storage data processing is relatively more efficient, in the later stages, as the bzz price retraces its fall, it cannot avoid the mass shift of storage service providers and cannot guarantee Swarm’s storage volume, while Filecoin guarantees the continuous growth of storage volume through constraints in various aspects.

The secret competition between Swarm and Filecoin’s functional value

  • Swarm – The future will serve the whole ecology through the application

Swarm provides content distribution service for Ether and DApp with the following functions.

  1. Swarm is a decentralized storage platform that provides a local base layer service for the Ethernet web3 stack.
  2. Swarm aims to replace the Ethernet on-chain storage solution and become a decentralized storage platform for Ethernet public records.

3, Swarm can assist DApps to store and distribute code, data and content without interfering with information on the blockchain.

Swarm is a decentralized storage and communication system used to serve the development of the digital Internet. Swarm is primarily a peer-to-peer storage and service solution for the Ethernet blockchain and for the development of the Internet with anti-DDoS aggressiveness, zero downtime fault tolerance and strong anti-censorship and self-sustainability features.

The four core components of the Swarm project are confirmed through the Swarm Orange Book: first, the underlying P2P network; second, the distributed storage service; third, the provision of high-level data access as the underlying functional definition API; and fourth, the application layer of the program based on the Ethernet blockchain network. the Swarm project makes full use of the different technologies and protocols of Ethernet.

Swarm project looks at the whole Internet ecosystem, Swarm still belongs to the underlying data storage network system, and will eventually also serve the whole ecosystem of the Internet by developing all kinds of decentralized applications.

  • Filecoin – Long-term value lies in the volume of data

In the long run, Filecoin is different from other blockchain projects in that it is not just used as a distributed storage, but is more of a data volume representative.

After Filecoin’s main online launch, the consensus in the market has shown a rising trend. A large number of development business level, launching all kinds of distributed storage-related products, its ecological applications, basic arithmetic, technical iterations, joint layouts, financial tools, etc. are in full swing. Both protocol labs, node participants and ecological developers are continuously enriching the Filecoin network so that the network of distributed storage can store real and valid data as early and as much as possible.

The main Filecoin application layer features are as follows.

Resilient data backup: Filecoin offers customers multiple options for backing up their data. Customers can back up their data to different nodes without limit and can verify the storage status of any node at any time. The Filecoin storage protocol is more resilient than the more costly decentralized backup and validation of centralized storage.
Off-chain data transfer: Filecoin provides a mechanism for off-chain data transfer. Customers can store larger files on their hard disks and use the off-chain method to place the hard disks directly on the storage server of the storage provider and continuously verify that the files are stored securely. This mechanism can solve the problem of slow and expensive uploading of large files.
As the infrastructure of Web3.0 data storage, Filecoin has the advantages of decentralization, fast transmission speed, high efficiency and security in the data storage market. With the landing of basic applications and the improvement of ecology, Filecoin will achieve a larger scale of application landing.

Conclusion.

Both Filecoin and Swarm are laying the foundation for the final form of distributed storage and contributing to the realization of the huge ecosystem and business value of distributed storage.

Posted by:CoinYuppie,Reprinted with attribution to:https://coinyuppie.com/an-article-detailing-the-head-to-head-battle-between-dark-horse-swarm-and-frontrunner-filecoin/
Coinyuppie is an open information publishing platform, all information provided is not related to the views and positions of coinyuppie, and does not constitute any investment and financial advice. Users are expected to carefully screen and prevent risks.

Like (1)
Donate Buy me a coffee Buy me a coffee
Previous 2021-05-31 09:07
Next 2021-05-31 09:42

Related articles