Stay organized with collections
Save and categorize content based on your preferences.
In June 2022, NVIDIA announced the upcoming end of support (EOS) of the
NVIDIA K80 GPU model. For Google Cloud, NVIDIA K80 GPUs will reach EOS
on May 1, 2024.
What do I need to know
As of May 1, 2024, you won't be able to create or access any VMs that are
running NVIDIA K80 GPUs.
How will this EOS impact my existing virtual machine (VM) instances
Until May 1, 2024, your VMs that are running NVIDIA K80 GPUs are not affected.
However, your projects might be at risk because you are running a GPU model
that is approaching its EOS. EOS products or features are not supported by
Google Cloud.
What do I need to do
For VMs that have attached NVIDIA K80 GPUs, select another
GPU model. We recommend the NVIDIA T4 GPUs.
Make a note of the machine type that is supported for the selected GPU model.
Modify the existing VM: This method is only supported if the
new GPU model is available in the same zone, and uses the N1 general-
purpose machine type.
Move to a new VM
You must move to a new VM if the new GPU model is not supported in the
zone of the existing VM, or if you are switching to a GPU that runs on an
accelerator-optimized machine type.
To move to a new VM, complete the following steps:
If your existing VM uses local SSD disks that contain data you want to keep,
move the contents of those disks to a persistent disk.
Move persistent disks from the old VM to the new VM. You can do this by
detaching the persistent disk from the old VM and adding it to the new VM.
You can alternatively
transfer files from one VM to the
other.
Optional: Move the saved data from persistent disks back to a local SSD.
If the new GPU model that you chose is supported in the same zone, and uses
the N1 general-purpose machine type, you can modify your existing VM to switch
from an NVIDIA K80 to the new GPU model.
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Hard to understand","hardToUnderstand","thumb-down"],["Incorrect information or sample code","incorrectInformationOrSampleCode","thumb-down"],["Missing the information/samples I need","missingTheInformationSamplesINeed","thumb-down"],["Other","otherDown","thumb-down"]],["Last updated 2025-08-29 UTC."],[[["\u003cp\u003eNVIDIA K80 GPUs will reach end of support (EOS) on Google Cloud on May 1, 2024, meaning you will not be able to create or access VMs running them after this date.\u003c/p\u003e\n"],["\u003cp\u003eWhile existing VMs with NVIDIA K80 GPUs are unaffected until May 1, 2024, it's crucial to migrate away from this GPU model as EOS products are not supported by Google Cloud.\u003c/p\u003e\n"],["\u003cp\u003eGoogle recommends switching to NVIDIA T4 GPUs and you should review supported machine types, regions, and zones for the new GPU.\u003c/p\u003e\n"],["\u003cp\u003eYou can either move to a new VM, which is required if the new GPU model is not supported in the current zone, or modify your existing VM if the new GPU is available in the same zone and uses the N1 general-purpose machine type.\u003c/p\u003e\n"]]],[],null,["# NVIDIA K80 end of support\n\n*** ** * ** ***\n\nIn June 2022, NVIDIA announced the upcoming end of support (EOS) of the\nNVIDIA K80 GPU model. For Google Cloud, NVIDIA K80 GPUs will reach EOS\non May 1, 2024.\n\nWhat do I need to know\n----------------------\n\nAs of May 1, 2024, you won't be able to create or access any VMs that are\nrunning NVIDIA K80 GPUs.\n\nHow will this EOS impact my existing virtual machine (VM) instances\n-------------------------------------------------------------------\n\nUntil May 1, 2024, your VMs that are running NVIDIA K80 GPUs are not affected.\n\nHowever, your projects might be at risk because you are running a GPU model\nthat is approaching its EOS. EOS products or features are not supported by\nGoogle Cloud.\n\nWhat do I need to do\n--------------------\n\n1. For VMs that have attached NVIDIA K80 GPUs, select another [GPU model](/compute/docs/gpus). **We recommend the NVIDIA T4 GPUs**.\n2. Make a note of the machine type that is supported for the selected GPU model.\n3. Review the available [GPU regions and zones](/compute/docs/gpus/gpu-regions-zones) for that GPU model.\n4. Complete *one* of the following:\n\n - [Move to a new VM](#move).\n - [Modify the existing VM](#modify): This method is only supported if the new GPU model is available in the same zone, and uses the N1 general- purpose machine type.\n\n### Move to a new VM\n\nYou must move to a new VM if the new GPU model is not supported in the\nzone of the existing VM, or if you are switching to a GPU that runs on an\n[accelerator-optimized machine type](/compute/docs/accelerator-optimized-machines).\n\nTo move to a new VM, complete the following steps:\n\n1. If your existing VM uses local SSD disks that contain data you want to keep,\n move the contents of those disks to a persistent disk.\n\n2. [Create a new VM](/compute/docs/gpus/create-vm-with-gpus) and specify your\n new GPU model.\n\n | **Note:** To reuse most of your configurations from your existing VMs, you can use the [create similar](/compute/docs/instances/create-vm-from-similar-instance) option. However, remember to update the GPU type.\n3. Move persistent disks from the old VM to the new VM. You can do this by\n detaching the persistent disk from the old VM and adding it to the new VM.\n You can alternatively\n [transfer files](/compute/docs/instances/transfer-files) from one VM to the\n other.\n\n4. Optional: Move the saved data from persistent disks back to a local SSD.\n\n5. [Reassign any static IP addresses](/compute/docs/ip-addresses/configure-static-external-ip-address#IP_assign)\n associated with the original VM to the new VM.\n\n6. [Install GPU drivers](/compute/docs/gpus/install-drivers-gpu) on the VM.\n\n7. Install your applications on the new VM.\n\n8. Delete the old VM.\n\n### Modify your existing VM\n\nIf the new GPU model that you chose is supported in the same zone, and uses\nthe N1 general-purpose machine type, you can modify your existing VM to switch\nfrom an NVIDIA K80 to the new GPU model.\n\n1. [Check that you have enough quota](/compute/docs/gpus/create-vm-with-gpus#check-quota) for the new GPU type.\n2. Verify that all of your critical applications are stopped on the VM.\n\n3. In the Google Cloud console, go to the **VM instances** page to see your list of\n VMs.\n\n [Go to VM instances](https://console.cloud.google.com/compute/instances)\n4. Click the name of the VM that is running the NVIDIA K80 GPU. The\n **VM instance details** page opens.\n\n5. Complete the following steps from the **VM instance details** page.\n\n 1. Click stop **Stop** . If there is\n no **Stop** option, click\n more_vert **More actions \\\u003e** stop **Stop**.\n Check the notification panel to see when the VM is stopped.\n\n 2. Click edit **Edit** and then\n complete the following steps:\n\n 1. In the **Machine configuration** section, select the **GPU** machine family.\n 2. From the **GPU type** drop-down, change the GPU model from NVIDIA K80 to your new GPU model.\n6. To apply your changes, click **Save**.\n\n7. To restart the VM, click **Start/Resume**.\n\nWhat's next\n-----------\n\n- Learn more [About GPUs](/compute/docs/gpus/about-gpus).\n- Review [GPU pricing](/compute/gpus-pricing)."]]
