FileStar, the decentralized Web3 storage infrastructure

FileStar

Decentralized Web3 storage infrastructure

PixelPlex has designed an incentive layer for a distributed storage network based on the IPFS protocol. The solution has been created on top of the Filecoin concept and comes with several technical improvements and innovations such as advanced hash algorithms, the removal of a pledged collateral, and recursive zk-SNARK technology.

Project goals

1
Facilitate an optimized utilization of computation, storage, and bandwidth resources
2
Help lower the threshold for mining
3
Solve the TPS and on-chain message congestion
4
Improve scalability and mining efficiency
5
Ensure a fair token launch and community governance
6
Optimize the performance of the existing Filecoin mining software
7
Lay the groundwork for a fully-fledged Web3 internet infrastructure

Work done

  • Design and implementation of an incentive layer for a distributed storage network

  • Modification of the WindowPoST spot check logic

  • Introduction of a recursive storage proof to enhance on-chain processing and TPS

  • Android/ iOS FileStar mobile wallet

  • Integration of the EVM for smart contract development

  • Creation of a sustainable and fair token governance system

Technologies used

  • echo icon
    Go
  • swift icon
    IPFS
  • health-kit icon
    Drand
  • core location icon
    IPLD
  • core location icon
    EVM
  • smart contaract icon
    libp2p

Business opportunity

FileStar is premised on the idea of Filecoin. The latter is a peer-to-peer network intended for storing files. It incorporates special built-in economic incentives to ensure that files are kept reliably over time.

In spite of its ground-breaking vision and great aspirations, Filecoin does not boast a polished design and sophisticated technical features that could help it demonstrate a greater performance.

On top of that, Filecoin’s project development and management are too centralized and its native token distribution and release are relatively unfriendly to miners. Among other obstacles that prevent it from gaining wider traction are the high threshold for participating in Filecoin mining and a limited on-chain processing capacity.

Space view of three planets

Solution

Inspired by Filecoin’s innovative idea, our PixelPlex team designed an incentive layer for decentralized storage called FileStar that can help usher in the Web3 internet infrastructure.

The solution is based on the same vision as Filecoin, yet comes with several improvements such as the removal of an initial pledged collateral, integration of the EVM, a more elaborate tokenomics vision, innovative hash algorithms, and much more.

FileStar is also slated to tackle the most pressing challenges of Filecoin and attract more hardware, providing abundant storage, bandwidth, and computation resources.

The UI of the transfer page of FileStar

Details

Taking into account the imperfections of the Filecoin project, our PixelPlex team has worked out and incorporated a set of special features in the FileStar solution. The sophisticated characteristics contribute to FileStar’s high performance, enable incentive at a more refined granular level, and facilitate an optimized utilization of computation, storage, and bandwidth resources.

The collage of a few UIs of FileStar

Innovative hash algorithms

fileCoin
Filecoin
Traditionally, Filecoin has used SHA256 algorithm that limits hardware compatibility and excludes the participation of a large number of Intel miners, which consequently prevents the Filecoin network’s growth.
VS
fileStar
FileStar
FileStar, on the other hand, encourages more diverse hardware into the network. That is why our team has proposed the implementation of other sophisticated algorithms, including SHA512, Poseidon, Pederson, and BLAKE2s. These all provide support to both Intel and other CPU mining architectures and facilitate network growth.

Recursive zk-SNARK technology

fileCoin
Filecoin

Proof of Copy (PoRep) is an essential part of the Filecoin storage proof system: when a miner seals one sector, PoRep requires them to submit two proofs to the network.

However, the on-chain message processing capability (TPS) of the Filecoin network is minimal. When the network becomes congested, proof messages occupy the majority of on-chain resources, and ordinary messages will not be packaged.

VS
fileStar
FileStar

FileStar proposes a recursive zk-SNARK technology to reduce the number of messages for submitting proofs, thus solving the TPS problem.

The basic concept of recursive zk-SNARK is that the proof of multiple sectors produced within a certain period will be verified off-chain, generating an aggregate proof. At the end of that period, only one proof needs to be submitted to the network, containing proof of multiple sectors.

The recursive zk-SNARK technology significantly improves the TPS and achieves greater network scaling. By adjusting the number of aggregated proofs, it will adapt the network’s processing capability to meet the needs of different development stages of the FileStar network in the future.

WindowPoST and VRF

fileCoin
Filecoin

In Filecoin, miners need to provide a Proof of SpaceTime (PoST) upon the completion of the PoRep so as to prove that the data has been continuously stored. If a miner fails to submit Proofs-of-Spacetime to the chain, they will be penalized.

The problem is that as the network grows, the number of WindowPoST submissions will also increase, which will lead to network congestion and affect message processing.

VS
fileStar
FileStar
FileStar offers a random checkup for WindowPoST and adopts a verifiable random function (VRF) to enhance the security of the random check, which helps lower the WindowPoST submission frequency. Plus, with a VRF, FileStar prevents any malicious activity coming from miners by not enabling them to predict the time of a checkup.

High-performing mining software

fileCoin
Filecoin
Filecoin’s mining software has many flaws in task scheduling, which has affected the network storage growth. The platform also badly needs optimization in sealing efficiency.
VS
fileStar
FileStar
Our experts proposed to release mining software with a refined task scheduling to increase the mining efficiency of the miner. The optimization mainly focuses on the task scheduling module and the zero-knowledge proof module.

Elaborate tokenomics model

To encourage miners to contribute to the storage, computing, and bandwidth of the project, we have designed a FileStar native token called STAR.

STAR can be used for paying for gas and storage and gives its holders the unbreakable right to vote on key decisions and further development of the FileStar project.

The UI of the transaction details page of FileStar

Decentralized governance

The FileStar community consists of developers, miners, and average token holders. Each of them can participate in the vote and jointly determine the network’s direction and growth as well as the development and launch of new features.

In the community, anyone has the right to submit a pool request with a complete test code. All submissions will be merged into the test network after a thorough analysis and will be launched on the mainnet once they have demonstrated adequate running.

A scheme of blocks symbolizing decentralized governance paradigm

EVM integration

Our developers have successfully integrated the Ethereum Virtual Machine (EVM) into FileStar. This allows the FileStar team to develop and deploy Ethereum-powered smart contracts right on the platform with minimal changes in the code.

Visualization of a blue EVM that reads binary code

Project features

A blue icon of an infographic

Lower mining requirements

A blue icon of scalability

High scalability

A blue icon of a data list

Great data availability

A blue icon of a decentralized scheme

Decentralized governance

A blue icon of a gear

Fair token distribution mechanism

A blue icon of a gear on a mobile display

Improved performance of mining software

A blue icon of a shield with a checkmark

Efficient use of resources

A blue icon of a line chart

Refined design of the economic model

Ongoing development and future plans

Our PixelPlex team firmly believes that the successful launch of FileStar’s incentive layer for decentralized storage is a significant achievement in the Web3 space.

We continue working on the enhancement of a proof mechanism for computing and bandwidth facilities to further attract different resource providers. This will allow us to create even more robust decentralized computing and distributed bandwidth networks.

As far as long-term goals are concerned, we are planning to:

  • enhance the smart contract fucntionality to be able to converge data chunks in more sophisticated storages
  • create a desktop application with a convenient interface that will help easily deploy a node and ensure its smooth communication with other nodes