start
  • ๐Ÿ‘‹Welcome
  • ๐Ÿ“–Introduction
  • ๐Ÿ’กUse Cases
  • ๐Ÿง‘Personas
    • Film Production
    • Animation Studios
    • Game Developer
    • Industrial Design
    • Advertising
    • AI Image Generation / Text-to-Image
    • Speech-to-Text & Text-to-Speech
    • AI Video Enhancement & Processing
    • AI Object Detection & Image Analysis
    • Enterprise LLM API
    • Private Knowledge Base LLM (RAG - Retrieval-Augmented Generation)
    • Family Photographer
    • Indie Game Developer
    • Aspiring 3D Artist
    • Playstation Gamer
  • ๐Ÿš€Get Started
    • Janction Node Operation Graphic Tutorial
  • ๐Ÿ”—Related Work
  • ๐Ÿ—๏ธArchitecture
    • Actor Model
  • ๐Ÿ–ฅ๏ธPooling
  • ๐Ÿช™Token
  • โšกColocation of Idle Processor Computing Power
  • โœ…Proof of Contribution
  • ๐ŸŽฎGPU Marketplace
    • Pricing strategy based on pvcg
  • โ“HELP FAQ
    • FAQ
      • How Janction Efficiently Stores AI/ML Models for Different Users๏ผŸ
      • Compared to traditional cloud GPU platforms, how does Janction's distributed idle GPU computing powe
      • How does Janction ensure the efficiency and quality of data annotation for various data types with d
      • How does Janction's execution layer handle the various AI subdomain functionalities?
      • How does Janction select and use different DAs?
      • Is Janction considering adopting the security guarantees provided by Restaking?
      • What is the current progress of Janctionโ€™s product technology?
      • How will Janction consider airdropping to the community?
  • ๐Ÿ›ฃ๏ธRoadmap
  • ๐Ÿ“œPolicy
    • Terms
Powered by GitBook
On this page

Architecture

PreviousRelated WorkNextActor Model

Last updated 11 months ago

The architecture of Janction mainly contains three parts Blockchain Layer, Distributed pooling of resources and GPU Marketplace.Blockchain Layer, as the settlement layer and DA layer of the distributed AI system, is mainly responsible for content data storage, transaction information storage and reward distribution. In the blockchain layer, data acquisition is done through browser extension crawling and on-chain data ingestion. Users can crawl their browsing data through Janction Extension, including Chatgpt question data, search engine browsing data, social APP data, etc., and then upload them to Janction Extension after user signature. Janction layer2 is mainly responsible for verifying the transaction process in the system and uploading it to the chain to record, the automatically running Proof of Contribution consensus algorithm will prove the contribution of the participants in the system and allocate rewards to the participants based on the contribution degree, and all the transaction processes in the chain are stored in the Data Availability layer. All transactions on the chain are stored in the Data Availability layer.

The core business processing part of Janction is Distributed pooling of resources, which pools GPU resources shared by users, and different types of GPUs are divided into vGPUs, each of which has equal arithmetic power and video memory, and are containerized to build a microservices architecture, which is composed of GPU pools through VxLan. The GPU pool is formed by VxLan, and when users need to use GPU resources, they will be allocated GPU resources through appropriate arithmetic routing and scheduling algorithms, and the rental behavior is priced by the PVCG mechanism.The AI microservice containers interoperate with each other through RESTful APIs, and Janction provides service api in the back-end to provide services for the users.

GPU marketplace is the front-end and back-end part of Janction system, which provides user personal information management, node management, service monitoring, system report and other services, including user reputation system in the user information, so as to evaluate the value of resources provided by the user and prevent malicious behavior.

๐Ÿ—๏ธ
Architecture