Aethir (ATH) Research Report

Project Name: Aethir
Tag: $ATH
Type: Decentralized GPU Network

Introduction

Aethir is a cloud computing infrastructure platform that provides users with a cloud platform built on a distributed GPU framework for sharing computing resources across different geographical locations and ownerships.

How It Works

Aethir breaks away from the traditional centralized model by deploying a distributed GPU framework, changing ownership, allocation, and usage patterns of enterprise-level GPUs. To maximize GPU utilization for high-demand tasks such as AI, machine learning (ML), and cloud gaming, Aethir achieves its goals through the following methods:

Distributed GPU Framework

Aethir’s cloud platform is built on a distributed GPU framework. This architecture connects independent GPU clusters into a unified network, increasing the cluster scale and improving reliability and resilience. By integrating geographically dispersed resources in this manner, Aethir handles large-scale AI workloads and reduces global dependency on major cloud providers.

Decentralized Ownership

The geographical dispersion of GPUs ensures distributed control over computing resources by the project team. Unlike traditional centralized cloud computing providers, the Aethir project team only handles the allocation and management of computing resources, while ownership remains with the providers of the mining rigs. Users providing computing resources can earn rental fees and project revenue shares, encouraging GPU miners to join and enhancing overall resource utilization.

Aethir Network

The Aethir network is driven by five fundamental roles: miners, developers, users, token holders, and the Aethir DAO. Miners are categorized into three types: containers, verifiers, and indexers. These roles collectively promote network consensus and work together to provide services.

Indexers

Indexers connect users with the most suitable containers, facilitating matching and ensuring quick startup of cloud-based applications and services. The goal is to provide “instant-on” services — transitioning from user requests to actual delivery in the shortest time possible.

Verifiers

Verifiers ensure the integrity and performance of containers within the Aethir network. They verify the container specifications provided by container providers to maintain the quality of network services.

Containers

Containers form the backbone of the Aethir network, serving as the actual usage points in the cloud. They act as virtual endpoints, executing and rendering applications (e.g., rendering games for players). By offloading workloads from local devices to containers (e.g., transferring all game execution and command processing), they provide users with an instantaneous and seamless cloud experience.

Economic Model

Aethir operates three main business models:

1. Cloud Gaming Model: To meet the demands of the evolving cloud gaming industry, Aethir provides computing resources optimized for gaming experiences.
2. AI Model: The Aethir platform offers a dedicated infrastructure for AI applications, enabling users to leverage the computing power required for complex AI model training and inference.
3. Virtualized Computing Model: Aethir also focuses on virtualized computing, providing flexible and scalable computing resources suitable for virtualized applications and workloads.
   
   Within the Aethir ecosystem, node operators have multiple revenue streams, divided into three reward forms:
   a. Service Fees: Buyers of computing power or demand-side entities pay service fees to purchase computing capacity. Payments are converted into ATH tokens, with 80% of the fees going to node operators, while Aethir retains a 20% platform share.
   b. Proof of Rendering Work: Token incentives are provided to node operators as an additional reward for completing computing tasks within the ecosystem. This encourages supply-side entities to join Aethir’s ecosystem and provide valuable processing and computing work. Proof of rendering work is distributed only after computing tasks are completed by containers.
   c. Proof of Capacity: Computing providers earn proof of rendering capacity by demonstrating their readiness to provide computing services. Even without active work, providers receive rewards to incentivize ecosystem participation, thereby reducing participation risk.

   Key Metrics

   

   Tokenomics

   Total Token Supply of 42,000,000,000:

• Airdrop: 6%
• Investors: 11.5%
• Verifiers and Compute Providers: 50%
• Team: 12.5%
• Advisors: 5%
• Ecosystem Fund: 15%