How to Install an nVidia GRID Card on Dell Servers

So, in the process of doing my job here at Dell as a VDI Support Guy in ProSupport, I periodically come across a customer issue where the GRID cards are not properly installed in the system, either because of a misconfiguration/oversight when ordering the server or just forgetting a step in the installation process.

After searching the Internet for some sort of doc or video that details the process and not finding anything, I stepped into our lab here in ProSupport and did an installation video.

Two key points:

  1. Make sure you ground yourself, protect your equipment from static discharge!!
  2. Don’t forget the power cable for the GRID.


Enjoy, and I hope this helps someone out. ¬†Please let me know if it does… below… in the comments.


5 thoughts on “How to Install an nVidia GRID Card on Dell Servers

  1. Jake

    Dell no longer directly sells the Grid K1 card – we are in the process of expanding our production clusters and would like to maintain the same config we have today (R730 with 1xK1 card).

    Do you see any pitfalls to us ordering an R730, specifying “with GPU cooling” in the config, and putting a K1 (obtained elsewhere of course) in ourselves? Do you know if there is a specific part number or model number of the GRID that would make it specific to being installed in a Dell?



    • scooterx01

      Hi Jake! Yeah, nVidia has discontinued the K-Series GPGPU cards and replaced them with the M-Series GRID 2.0 cards. Unfortunately, nVidia chose to start charging a licensing fee/per user for doing shared GPGPU. So, a lot of folks like you are wanting to deploy the K1/K2 models instead.
      The GPU cooling kit is primarily: low profile heat sinks for the cpu’s – so the dual-height GPGPU’s will fit in the server & requirement that 1100W power supplies are in the box to adequately cover the power needs of the system + GPGPU cards on board (plus a couple of other tweeks & riser power cable). I don’t see why the GPU Cooling config wouldn’t work, as a double-wide GPU is a double-wide GPU in terms of form factor.
      However, I would defer to your Dell Sales Rep to validate the config and ability to get the GPU Cooling config without purchasing a current M-Series GPGPU with the system.
      As for the what model to purchase; only the K1 & K2/K2a GPGPU’s were validated on the R720/R730 platforms, so stay within those bounds.
      Hope this helps,


      • Jake

        Thanks Scott – I noticed that some GRID K1’s appear to be OEM branded with unique model/mfg numbers. For example, if one were to look at a “RF61J” you would find it as a “Dell nVidia GRID K1” but it you were to look at a “J0G94A” you would find it as an “HP ProLiant nVidia GRID K1” – this is what threw me off.

        I assumed a K1 was a K1 – any idea what, if any difference there is? Or what the manufacture part number of the “Dell Certified” card is?

        Liked by 1 person

  2. scooterx01

    Hey Jake,
    Had to take a look into this. Apparently, nVidia leaves some small features/configurations available for resellers/hardware vendors to customize – either by menu option or through card firmware. So, therefore, each vendor will have unique model/mfg numbers and you should only apply the corresponding vendor’s firmware to that device.
    As for make/model numbers for Dell Certified cards – I did a quick search on the website. It seems we still have some available, even though they have been discontinued by nVidia.
    The mfg number for what we have available is: H6VT2


    – Scott


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s