> For the complete documentation index, see [llms.txt](https://docs.nuvolos.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.nuvolos.com/how-to-guides/workflows-for-students/working-with-a-gpu-in-a-course.md).

# Collaborate and use special resources

## Work in a group project

<mark style="color:$primary;">**Outcome**</mark>\
You collaborate with teammates inside a shared group Instance, editing the same files and saving snapshots together.

<mark style="color:$primary;">**Before you start**</mark>

* Your instructor has set up a group Instance for your group.
* You have accepted the invitation to the group Instance and can see the group Space in your Dashboard.

Group projects in Nuvolos use a **shared Instance** - one Instance that several students access as Instance Editors. All edits, files, and snapshots in the group Instance are visible to every group member in real time.

### **Start the shared Application**

{% stepper %}
{% step %}
From the Dashboard, open the group-work Space.
{% endstep %}

{% step %}
Navigate to your group Instance.
{% endstep %}

{% step %}
Open the Applications view and start the Application your instructor configured for group work.
{% endstep %}
{% endstepper %}

The startup procedure is identical to starting any other Application - see [Start an Application](/how-to-guides/common-workflows/starting-an-application.md) above.

### Edit together without conflicts

Two cases to be aware of:

* **If the Application runs in Shared mode** (most commonly JupyterLab 4.0.0 or later) - you and your teammates can edit the same file simultaneously and see each other's changes live, similar to Google Docs. Edits merge automatically, with no file-locking and no version conflicts.
* **If the Application does not run in Shared mode** - multiple people editing the same file at the same time produces application-dependent results. Sometimes one set of changes wins, sometimes the file is corrupted. Coordinate before editing: break work into separate files where possible, agree on who edits what, and communicate through your group's chat.

{% hint style="info" %}
You cannot tell from the student-facing UI whether Shared mode is enabled - it is configured by your instructor. If your group has not been told, assume non-shared mode and coordinate before editing the same file.
{% endhint %}

### Save the group's progress with snapshots

Snapshots in a group Instance work the same way as in your personal Instance, but they preserve everyone's work, not just yours. Take one at the end of each working session and before any major change. See [Take a snapshot](/how-to-guides/common-workflows/snapshots/create-a-snapshot.md) above.

{% hint style="info" %}
Pick a snapshot-naming convention with your group early - date-based names work well. Anyone in the group can restore a snapshot, which affects the entire Instance and therefore everyone, so coordinate before doing so.
{% endhint %}

## Work with a GPU in a course

<mark style="color:$primary;">**Outcome**</mark>\
You run a GPU-enabled Application during a scheduled lab session or on demand, depending on how your instructor has set up the course.

<mark style="color:$primary;">**Before you start**</mark>

* Your instructor has enabled GPU access for your course.
* You know which workflow your course uses - GPU Lab Sessions or On-Demand GPU.

Nuvolos supports two GPU workflows for courses, and the procedure differs significantly between them. Check with your instructor or the course README which one applies before continuing.

### GPU Lab Sessions

In Lab Session workflows, your instructor schedules specific times during which you have GPU access. You do not start the GPU-enabled Application yourself - the system starts it for you at the scheduled time.

Three rules to follow:

* **Sign up for the course at least one hour before the first scheduled session.** The system processes the attendee list well before each session start, and last-minute sign-ups will not get a GPU-enabled machine.
* **Do not start the Application manually.** If you have an Application running with a non-GPU size when the session begins, the system will restart it with GPU access a few minutes before the start. Any other action on your part is unnecessary.
* **Do not stop the GPU-enabled Application.** The system shuts down all student Applications at the end of the session. If you stop the Application yourself during the session, you will not be able to restart it.

### Confirm your Application has GPU enabled

Hover your mouse over the Application's icon on the left sidebar. A tooltip with a green badge and the GPU model (e.g. *"on Tesla T4 GPU"*) confirms GPU is active.

### **On-Demand GPU courses**

In On-Demand workflows, **you start the GPU-enabled Application yourself**. The Applications view shows three pieces of information you need to monitor:

* **Size column** - displays the GPU model attached to the Application (for example, Tesla T4). You cannot change the size yourself; if no GPU-enabled size is available, contact your instructor.
* **Credit/hour column** - how many Credits it costs to run this Application for one hour.
* **Credit progress bar (top right)** - how much Credit you have used so far against your limit, and the end date of that limit.

{% hint style="info" %}
Credit charges start when you click the start button, not when the web UI loads. To make the most of your Credits, only start the Application when you have at least 30 consecutive minutes to work, and always stop the Application when you no longer need it.
{% endhint %}

### **What happens at the Credit limit**

Each Credit limit has an end date. At midnight UTC on the end date, one of three things happens:

* Your Credit usage resets to zero and a new period with a new limit begins.
* Your Credit usage remains and a new period with a higher limit begins.
* No further limit period is defined - *Credit limit* and *Ends on* show *N/A*. In this case, you cannot run GPU Applications anymore.

Two things to be aware of:

* **Reaching your limit stops all running Applications with a non-zero Credit price.** You must wait for the next limit period to continue.
* **On end dates, running Applications are stopped at midnight UTC** if Credit usage is configured to reset to zero.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## 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://docs.nuvolos.com/how-to-guides/workflows-for-students/working-with-a-gpu-in-a-course.md?ask=<question>
```

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.