Introduction

EMETH GPU POOL is a dynamic cloud computing platform designed to minimize the costs and complexities associated with compute-intensive tasks. This platform facilitates access to a wide array of GPU resources, allowing users to capitalize on available computing power efficiently. Our software allows anyone to easily become a host and start earning income with their idle machine resources. Additionally, our powerful search tool helps users swiftly locate the most suitable and cost-effective compute solutions tailored to their precise needs.
Key Functions
  • Launch powerful GPU instances at the best price in mere seconds.
  • Select from flexible rental options, including on-demand, interruptible and reserved 1M-3M-1Y
  • Discover optimal deals that align precisely with your compute needs.
  • Choose from pre-built templates for effortless one-click setup.
  • Automatic SSH and Jupyter configurations are available for most standard images.
  • Benefit from persistent storage solutions.
  • Access our platform through a user-friendly GUI.
  • Opt for your preferred level of security.
  • Experience the most competitive GPU instance pricing available globally.
How it works

Our platform simplifies and automates the hosting process, giving providers complete autonomy to set their prices and specify rental contract terms. Our user-friendly interface enables precise searches for machines, filtering by crucial specifications such as GPU type, GPU RAM, CPU, system RAM, connectivity, and more. This detailed approach ensures that users can find the exact computing resources they need for their projects.

Talk To Us

If you encounter any issues or need further assistance, please contact our support team at support.emeth-gpu-pool@alt.ai. We're here to help!

Quick Start

  1. Register an Account & Add Credit

    Before you can rent a machine from EMETH GPU POOL, you will need to first register an account and add credit. If you already have an account, simply click the Login button to continue. If you are new to the platform, click on the Create Account link to establish a username and password.

    Click on the Add Credit button on the Billing page to add credits to your account. If your balance runs low, your instances will be stopped until you add more credits. To avoid interruptions, you can set up an auto-payment threshold on the Billing page. This setting automatically charges your card when your credit balance falls below the specified threshold.

  2. How to connect to your instances

    To connect to your instances, it's important to configure your account and device based on your preferred method of connection. For SSH access, you'll need to generate an SSH key and upload the public key to EMETH GPU POOL.

  3. Filter function

    Our user-friendly interface enables precise searches for machines, filtering by crucial specifications such as GPU type, GPU RAM, CPU, system RAM, connectivity, and more. This detailed approach ensures that users can find the exact computing resources they need for their projects.

  4. Select a template

    Most templates support built-in Jupyter and SSH options for enhanced flexibility. See the template list and select the one that best fits your use case, then simply click the Deploy button to get started.

  5. Rent

    Enter the instance name, specify the desired resources, and then click on the Confirm Renting button to accept the offer and rent an instance. Once rented, you can view the instance on the Your Instances page.

  6. How to work after renting an instance
    • Connect: Connect your instances.
    • Start: After stopping, you can start your instances.
    • Stop: Stop your instances.
    • Reboot: Restart your instances.
    • Backup/Restore: Backup/ Restore your instances if you need.
    • Terminate: Stop running your instances.

Make sure to destroy instances once you've finished using them. Stopping the instance will prevent GPU charges. Stopped instances can be restarted when the GPU becomes available again. Properly managing your instances helps optimize costs and resources.

Hosting

EMETH GPU POOL serves as a dynamic GPU marketplace, empowering individuals to effortlessly become hosts and generate income from their idle machine resources. With our software, users can seamlessly enter the hosting space, maximizing the value of their hardware assets while contributing to a vibrant community of computing resources.

Hosts are responsible for:

  • Account setup and hosting agreement: You must create a new account for hosting. Once your account is created, to proceed, take a moment to read and agree to the EMETH GPU POOL Hosting Terms of Use. After you submitted your agreement, you can start to deploy your machines from the Setup page.
  • Installing Ubuntu: Begin by installing the Ubuntu operating system on your machine.
  • Creating Disk Partitions: Partition your disk according to your requirements, ensuring adequate space for your system and data.
  • Installing NVIDIA Drivers: Install the appropriate NVIDIA drivers to enable GPU functionality on your system.
  • Opening Network Ports on the Router: Configure your router to open the necessary network ports required for communication with external services.
  • Installing Hosting Software: Install the hosting software to enable hosting and rental services for your GPUs.
  • Testing and Troubleshooting: Conduct thorough testing and troubleshooting to address any issues that may arise, such as driver conflicts, errors, malfunctioning GPUs, or network port issues.
  • No Support for Machine Setup: Please note that we do not offer support for setting up your machine. You are responsible for ensuring that your system is configured correctly.
  • Managing Listings and GPU Offers: Take charge of managing your listings and GPU offers for rentals. This includes setting pricing, specifying end dates for offers, and ensuring accurate and up-to-date information for potential renters.
  • Planning Maintenance: Plan for routine maintenance to minimize disruptions to client jobs. Ensure that maintenance activities are scheduled in a way that avoids impacting ongoing tasks.