In diesem Dokument wird beschrieben, wie Sie eine Standardlaufzeit mit GPUs für Colab Enterprise-Nutzer in einem Projekt aktivieren. Nachdem Sie eine Standardlaufzeit mit GPUs aktiviert haben, können Nutzer in ihrem Notebook per Mausklick von einer regulären Standardlaufzeit zu einer Standardlaufzeit mit GPUs wechseln.
Dieses Dokument richtet sich an Colab Enterprise-Administratoren, die Standardlaufzeiten mit GPUs für andere Nutzer in ihrer Organisation aktivieren möchten. Dabei wird davon ausgegangen, dass Sie über Kenntnisse zu folgenden Themen verfügen:
So verwalten Sie Colab Enterprise-Laufzeiten und ‑Laufzeitvorlagen.
Verwenden von Identity and Access Management (IAM) zur Zugriffssteuerung
Übersicht
Damit Nutzer zu einer Standardlaufzeit mit GPUs wechseln können, muss ein Administrator (roles/aiplatform.colabEnterpriseAdmin) oder ein Nutzerkonto mit der Berechtigung aiplatform.notebookRuntimeTemplates.create zuerst eine Standardlaufzeit mit GPUs erstellen.
Wenn Sie zum ersten Mal eine Standardlaufzeit mit GPUs erstellen, wird in Colab Enterprise eine neue Standardlaufzeitvorlage mit GPUs in den Spezifikationen erstellt. Die ursprüngliche Standardlaufzeit ist davon nicht betroffen und bleibt bestehen, bis sie abläuft oder gelöscht wird. Nachdem die neue Standardlaufzeitvorlage mit GPUs erstellt wurde, kann jeder Nutzer mit der Berechtigung aiplatform.notebookRuntimes.assign für das Projekt und der Berechtigung aiplatform.notebookRuntimeTemplates.apply für die Laufzeitvorlage eine Standardlaufzeit mit GPUs erstellen und verwenden. Diese Berechtigungen sind in der Rolle „Colab Enterprise User“ (roles/aiplatform.colabEnterpriseUser) enthalten.
Spezifikationen
Die Standardlaufzeit mit GPUs hat andere Spezifikationen als die ursprüngliche Standardlaufzeit. Die verfügbaren Maschinentypen, GPUs und Datenträgertypen variieren je nach Region. Daher können einige Spezifikationen von der ursprünglichen Standardlaufzeit abweichen.
In der folgenden Tabelle werden die Spezifikationen für eine Standardlaufzeit mit GPUs basierend auf der Region der Standardlaufzeit beschrieben.
Beschreibung der Region
Spezifikationen für Standardlaufzeit
Regionen mit Unterstützung für L4-GPUs
Maschinentyp: g2-standard-4
Beschleuniger: 1 NVIDIA_L4-Beschleuniger
Datenlaufwerk: 100 GB pd-balanced
Regionen, die keine L4-GPUs, aber T4-GPUs unterstützen
Maschinentyp: n1-standard-4
Beschleuniger: 1 NVIDIA_TESLA_T4-Beschleuniger
Datenlaufwerk: 100 GB pd-standard
Regionen, die keine L4- oder T4-GPUs unterstützen
Standardlaufzeiten mit GPUs werden nicht unterstützt.
Verfügbarkeit von Beschleunigern
Colab Enterprise unterstützt Standardlaufzeiten mit den folgenden Beschleunigertypen:
L4
T4
Informationen zur regionalen Verfügbarkeit dieser Beschleuniger finden Sie unter Beschleuniger verwenden.
Hinweise
Sign in to your Google Cloud account. If you're new to
Google Cloud,
create an account to evaluate how our products perform in
real-world scenarios. New customers also get $300 in free credits to
run, test, and deploy workloads.
In the Google Cloud console, on the project selector page,
select or create a Google Cloud project.
Bitten Sie Ihren Administrator, Ihrem Nutzerkonto die IAM-Rolle „Colab Enterprise Admin“ (roles/aiplatform.colabEnterpriseAdmin)
IAM role on the project.
In the Region menu, select the region that contains your notebook.
Click the notebook that you want to open. If you haven't created a notebook yet,
create a notebook.
In your notebook, click Connect.
After Colab Enterprise connects to the default runtime, in the
top right corner of your notebook, click the button to switch to
a default runtime with GPUs. For example, if your notebook is in
a region that supports L4 accelerators, click
Switch to L4.
Colab Enterprise creates a new default runtime that has
GPUs, and then connects to the runtime. The ability to switch to a
default runtime with GPUs is enabled for other users in the project.
Turn off GPUs for default runtimes
To turn off the ability to switch to a default runtime with GPUs, you
must delete the runtime template named Default with GPU.
See Delete a runtime template.
Limitations
Default runtimes with GPUs are only available in regions that support
specific accelerator availability. See
Accelerator availability.
You must first connect a notebook to a default runtime to be able to switch to
a default runtime with GPUs.
[[["Leicht verständlich","easyToUnderstand","thumb-up"],["Mein Problem wurde gelöst","solvedMyProblem","thumb-up"],["Sonstiges","otherUp","thumb-up"]],[["Schwer verständlich","hardToUnderstand","thumb-down"],["Informationen oder Beispielcode falsch","incorrectInformationOrSampleCode","thumb-down"],["Benötigte Informationen/Beispiele nicht gefunden","missingTheInformationSamplesINeed","thumb-down"],["Problem mit der Übersetzung","translationIssue","thumb-down"],["Sonstiges","otherDown","thumb-down"]],["Zuletzt aktualisiert: 2025-08-19 (UTC)."],[],[],null,["# Enable default runtimes with GPUs\n=================================\n\n|\n| **Preview**\n|\n|\n| This feature is subject to the \"Pre-GA Offerings Terms\" in the General Service Terms section\n| of the [Service Specific Terms](/terms/service-terms#1).\n|\n| Pre-GA features are available \"as is\" and might have limited support.\n|\n| For more information, see the\n| [launch stage descriptions](/products#product-launch-stages).\n\nFor support during the preview, email\n[vertex-notebooks-previews-external@google.com](mailto:vertex-notebooks-previews-external@google.com).\n\nThis document describes how to enable a default runtime with GPUs for\nColab Enterprise users in a project. After enabling a\ndefault runtime with GPUs, users can switch from a regular default runtime\nto a default runtime with GPUs by clicking a button in their notebook.\n\nThis document is intended for Colab Enterprise administrators who\nwant to enable default runtimes with GPUs for other users in their\norganization. It assumes you have knowledge of the following:\n\n- How to manage Colab Enterprise runtimes and runtime templates.\n- How to use Identity and Access Management (IAM) to control access.\n\nOverview\n--------\n\nTo enable users to switch to a default runtime with GPUs, an administrator\n([`roles/aiplatform.colabEnterpriseAdmin`](/colab/docs/access-control#aiplatform.colabEnterpriseAdmin))\nor a user account with the `aiplatform.notebookRuntimeTemplates.create`\npermission must first create a default runtime with GPUs.\n\nThe first time that you create a default runtime with GPUs,\nColab Enterprise creates a new default runtime template that\nincludes GPUs in its specifications. The original default runtime isn't\naffected and exists until it expires or it's deleted. After the\nnew default runtime template with GPUs is created, any user with\nthe `aiplatform.notebookRuntimes.assign` permission on the project\nand the `aiplatform.notebookRuntimeTemplates.apply` permission on the\nruntime template can create and use a default runtime with GPUs. These\npermissions are included in the Colab Enterprise User\n([`roles/aiplatform.colabEnterpriseUser`](/colab/docs/access-control#aiplatform.colabEnterpriseUser))\nrole.\n\n### Specifications\n\nThe default runtime with GPUs has different specifications than\nthe original default runtime. The machine types, GPUs, and data disk types\nthat are available vary by region, so some specifications can be different\nfrom your original default runtime.\n\nThe following table describes the specifications for\na default runtime with GPUs based on the region of the default runtime.\n\n### Accelerator availability\n\nColab Enterprise supports default runtimes with the following\naccelerator types:\n\n- L4\n- T4\n\nTo learn about the regional availability of these accelerators, see\n[Using accelerators](/vertex-ai/docs/general/locations#accelerators).\n\nBefore you begin\n----------------\n\n- Sign in to your Google Cloud account. If you're new to Google Cloud, [create an account](https://console.cloud.google.com/freetrial) to evaluate how our products perform in real-world scenarios. New customers also get $300 in free credits to run, test, and deploy workloads.\n- In the Google Cloud console, on the project selector page,\n select or create a Google Cloud project.\n\n | **Note**: If you don't plan to keep the resources that you create in this procedure, create a project instead of selecting an existing project. After you finish these steps, you can delete the project, removing all resources associated with the project.\n\n [Go to project selector](https://console.cloud.google.com/projectselector2/home/dashboard)\n-\n [Verify that billing is enabled for your Google Cloud project](/billing/docs/how-to/verify-billing-enabled#confirm_billing_is_enabled_on_a_project).\n\n-\n\n\n Enable the Vertex AI, Dataform, and Compute Engine APIs.\n\n\n [Enable the APIs](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com, dataform.googleapis.com, compute.googleapis.com&redirect=https://console.cloud.google.com)\n\n- In the Google Cloud console, on the project selector page,\n select or create a Google Cloud project.\n\n | **Note**: If you don't plan to keep the resources that you create in this procedure, create a project instead of selecting an existing project. After you finish these steps, you can delete the project, removing all resources associated with the project.\n\n [Go to project selector](https://console.cloud.google.com/projectselector2/home/dashboard)\n-\n [Verify that billing is enabled for your Google Cloud project](/billing/docs/how-to/verify-billing-enabled#confirm_billing_is_enabled_on_a_project).\n\n-\n\n\n Enable the Vertex AI, Dataform, and Compute Engine APIs.\n\n\n [Enable the APIs](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com, dataform.googleapis.com, compute.googleapis.com&redirect=https://console.cloud.google.com)\n\n### Required roles\n\n\nTo ensure that your user account has the necessary\npermissions to enable default runtimes with GPUs in Colab Enterprise,\n\nask your administrator to grant your user account the\n\n\nColab Enterprise Admin ([`roles/aiplatform.colabEnterpriseAdmin`](/vertex-ai/docs/general/access-control#aiplatform.colabEnterpriseAdmin)`)\nIAM role on the project.\n\n\n`\n| `\n| `**Important:**` You must grant this role\n| to your user account, `*not*` to your user account. Failure to grant the role to the correct principal might result in permission errors.\n| `\n`\n\n\nFor more information about granting roles, see `[Manage access to projects, folders, and organizations](/iam/docs/granting-changing-revoking-access)`.\n\n`\n\n\u003cbr /\u003e\n\n`\n\n\n`\n\n\nYour administrator might also be able to give your user account\nthe required permissions through [custom\nroles](/iam/docs/creating-custom-roles) or other [predefined\nroles](/iam/docs/roles-overview#predefined).\n`\n\n\n` One or more of the required roles includes the `dataform.repositories.list` permission. Users who are granted the `dataform.repositories.list` permission or the [Code Creator\n| (`roles/dataform.codeCreator`)](/iam/docs/understanding-roles#dataform.codeCreator) role in a project can list the names of code assets in that project by using the Dataform API or the Dataform command-line interface (CLI). Non-administrators using BigQuery Studio can only see code assets that they created or that were shared with them.\n`\n\n\n`\n\nEnable GPUs for default runtimes\n--------------------------------\n\n`\n\n\n`\n\nTo enable GPUs for default runtimes, do the following:\n`\n\n\n`\n\n1. In the Google Cloud console, go to\n the Colab Enterprise **My notebooks** page.\n\n [Go to My notebooks](https://console.cloud.google.com/vertex-ai/colab/notebooks)\n2. In the **Region** menu, select the region that contains your notebook.\n\n3. Click the notebook that you want to open. If you haven't created a notebook yet,\n [create a notebook](/colab/docs/create-console-quickstart#create).\n\n4. In your notebook, click **Connect** . \n\n\n5. After Colab Enterprise connects to the default runtime, in the\n top right corner of your notebook, click the button to switch to\n a default runtime with GPUs. For example, if your notebook is in\n a region that supports L4 accelerators, click\n **Switch to L4** . \n\n\n`\n\n\n`\n\nColab Enterprise creates a new default runtime that has\nGPUs, and then connects to the runtime. The ability to switch to a\ndefault runtime with GPUs is enabled for other users in the project.\n`\n\n\n`\n\nTurn off GPUs for default runtimes\n----------------------------------\n\n`\n\n`\n\nTo turn off the ability to switch to a default runtime with GPUs, you\nmust delete the runtime template named `Default with GPU`.\nSee [Delete a runtime template](/colab/docs/create-runtime-template#delete).\n`\n\n`\n\nLimitations\n-----------\n\n`\n\n`\n\n- Default runtimes with GPUs are only available in regions that support\n specific accelerator availability. See\n [Accelerator availability](#accelerator-availability).\n\n- You must first connect a notebook to a default runtime to be able to switch to\n a default runtime with GPUs.\n\n`\n\n`\n\nWhat's next\n-----------\n\n`\n\n`\n\n- To manage your runtime, see [Manage runtimes](/colab/docs/manage-runtimes).\n- Learn more about [runtimes and runtime templates](/colab/docs/runtimes).\n\n`\n\n\n``\n\n`"]]
