# Rent GPUs {% hint style="warning" %} Please note that NetMind is a distributed platform where all machines are contributed by users. We prioritize machines with high stability, but the Power platform cannot guarantee the uptime of the machines. In the event of machine downtime, we will fully refund the fees for the day of downtime. When using the machine, please be sure to back up your data in a timely manner to avoid data loss. {% endhint %} Follow these steps to rent a server with GPUs. ## Rent New GPUs {% hint style="warning" %} When you request a machine, you can specify a rental period of up to 30 days. We will send you a reminder to renew the lease when you have less than one day remaining. If you do not renew, upon expiry, the machine will be made available for others to rent, and all your data will be permanently and securely removed from the server. {% endhint %} 1. Click "Search GPU" on the sidebar under "GPU Clusters" to access the GPU rental feature. 2. In the "Search GPU" page, view and select the desired GPU resources, then click "Rent machine" to proceed to the machine setup page.

3. Complete your environment setup. 4. The system will calculate the fees to be paid in real time. 5. Users can choose between two rental programme: "On-demand" or "by day". Along with two payment methods: USD or NMT 6. Click "Make Payment" to enter the payment page.

### Pay by USD You need to add funds for USD before make payment.

*** ## Connect to Your Machine After you make the payment, you should now see a dashboard with your newly rented machine. If you do not see the IP Address and Port, you may need to manually refresh the browser page. You can copy the SSH command by clicking on the button as shown below.

The command will look like this ```bash ssh root@rent-gpus.netmind.ai -p {{SSH_PORT}} ``` If you're on Linux os MacOS, you will usually be able to pass this command directly to your terminal in order to connect. For Windows users, you can SSH using the PuTTY software from [this link](https://www.putty.org).\ Our product also supports IDE connection (for example via VSCode). {% hint style="info" %} If any failure or technical issue is detected on the server, the machine status on the dashboard might be showing as "Offline". In that case a "restart" button should appear, please restart the machine. If the problem persists, contact our support at . {% endhint %} *** ## Data Sync It is important to note that once the rental is terminated and the machine is released back to our pool of available servers, retrieving data becomes impossible. We recommend that you perform a backup of any relevant data, either locally or to the cloud, before terminating the rental. ### Local backup If you're on a Linux or MacOS system, this could be achieved via Secure Copy with the `scp` command. The example below shows how to donwnload the `/data/example.zip` folder from the remote machine to your local machine in the home directory, when run form your local machine. You will need to adjust the port and IP address accordingly. On Windows, you can use [WinSCP](https://winscp.net/eng/index.php). ```bash scp -P {{SSH_PORT}} root@rent-gpus.netmind.ai:/data/example.zip ~/example.zip ``` {% hint style="info" %} We only recommended using it for small files under 100MB. {% endhint %} ### Cloud Sync We encourage you to use cloud platforms for data synchronization services. We provide the examples of using Google Cloud, DropBox, and AWS as three cloud storage tools for data synchronization [here](/netmind-power-documentation/~/changes/88/rent-gpus/cloud-sync.md). *** ## Expose Your Service Since the environment itself does not have a public IP, you need to use an ingress service to access services within the environment from the internet. NetMind provide a free ingress service that supports proxying for up to one port. You can set the port you wish to expose in the custom template. Once the environment is created, you can view the corresponding address that is proxied to the internet in the "Rent GPUs detail" page.

{% hint style="info" %} Alternatively, you can choose to use third-party tools like **ngrok** to implement the ingress service. Here is an example to demonstrate how to set up the Ngrok as Ingress service [here](/netmind-power-documentation/~/changes/88/rent-gpus/ingress-service.md). {% endhint %} *** ## Machine status

Status	Explation
To Be Paid	The cost of the GPUs you have chosen to rent has been calculated. You need to complete payment before NetMind allocates the machine for you.
Queuing	NetMind is processing your payment and locating the machines for you.
Running	Your machine is now running. Connect to it via SSH.
Offline	The machine you rented has gone offline. NetMind is working to bring it back online.
Expired	The rental period for your machine with NetMind has expired.

--- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://netmind-power.gitbook.io/netmind-power-documentation/~/changes/88/rent-gpus.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.