👋Welcome

Janction is the first Layer2 to provide verifiable, synergic and scalable AI service. Janction uses smart contracts to automate and dynamize machine learning and artificial intelligence. Key elements like AI models, GPU computing power, data feeding, and data labeling all are integrated into Janction for coprocessing.

We aim to build a global distributed computing power market, where Janction GPU Marketplace aims to provide unlimited GPU capacity to users at lower costs by aggregating GPUs from multiple sources.

Janction will focus on the core design of the Janction cluster GPU pool and its billing system. First, let's analyze the Janction cluster GPU pool and its billing system. The cluster GPU pool is a system for managing and allocating multiple GPU resources, supporting large-scale parallel computing applications. It needs to consider resource scheduling and management, task scheduling, parallel computing, load balancing, resource security and isolation, fault recovery and fault tolerance, cost management and efficiency optimization, proof of work, and consistency issues, among others. There are two ways to allocate resources in the cluster GPU pool: centralized scheduling and P2P scheduling.