1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

NVIDIA Launches NVLINK - High-Speed GPU Interconnect

Discussion in 'Frontpage news' started by Hilbert Hagedoorn, Mar 25, 2014.

  1. Hilbert Hagedoorn

    Hilbert Hagedoorn Don Vito Corleone Staff Member

    Messages:
    35,489
    Likes Received:
    4,652
    GPU:
    AMD | NVIDIA
    NVIDIA announced that it plans to integrate a high-speed interconnect, called NVIDIA NVLink, into its future GPUs, enabling GPUs and CPUs to share data five to 12 times faster than they can today. Thi...

    NVIDIA Launches NVLINK - High-Speed GPU Interconnect
     
  2. Spets

    Spets Ancient Guru

    Messages:
    2,977
    Likes Received:
    39
    GPU:
    RTX 2080Ti XC Ultra
    Can't wait :D
     
  3. thatguy91

    thatguy91 Ancient Guru

    Messages:
    6,650
    Likes Received:
    97
    GPU:
    XFX RX 480 RS 4 GB
    It is talking about a new connection method, which means new motherboard slots, new chipsets etc., as well as the CPU support. It's intention isn't mainstream.
     
  4. schmidtbag

    schmidtbag Ancient Guru

    Messages:
    4,132
    Likes Received:
    1,167
    GPU:
    HIS R9 290
    Considering its called NVlink, without any definition as to what NV means, it's easy to assume that this is proprietary. What confuses me is if this actually has intended use with x86 - IBM doesn't support x86 and I haven't heard any sources saying intel or AMD will support it.

    But what I don't get is why this is necessary. As far as I'm aware, we haven't even saturated PCIe 2.0 yet (at least with modern single-GPU cards). I could understand this fixing latency issues, but I never got the impression latency was all that bad. PCIe should have plenty of bandwidth.

    I'm not against the idea of having a successor to AGP (PCIe is more of a successor to PCI), but I don't see nvidia getting very far with this in the PC market if they can't at least get intel on board with this.

    I hope both AMD and intel support it, because this (and DDR4 memory) make a good milestone for me to do a complete system upgrade. I only do full upgrades whenever there's a major system-wide change.
     

  5. ruiner13

    ruiner13 Member Guru

    Messages:
    145
    Likes Received:
    2
    GPU:
    Asus Strix 1080
    To me it sounds like they are putting an IBM Power CPU on the GPU card to act as a middle-ground between the main GPU functionality and the main CPU. Basically, offload as much as you can and make the GPU its own virtual machine. I am curious how they are going to accomplish the direct main memory access unless they have some new connector, or somehow the GPU will connect directly to the RAM slots.
     
  6. BLEH!

    BLEH! Ancient Guru

    Messages:
    5,860
    Likes Received:
    62
    GPU:
    Sapphire Fury
    And they intend to connect this how exactly?
     
  7. -Tj-

    -Tj- Ancient Guru

    Messages:
    16,259
    Likes Received:
    1,404
    GPU:
    Zotac GTX980Ti OC
    Probably something like nvidia g-sync style
     
    Last edited: Mar 26, 2014
  8. Ch0plol

    Ch0plol Member Guru

    Messages:
    167
    Likes Received:
    0
    GPU:
    ASUS GTX 770 4GB
    NV just means Nvidia

    Also, I'm fairly certain this is mostly intended for large datacenter / supercomputer applications. Not so much PCs.
     
  9. Lane

    Lane Ancient Guru

    Messages:
    6,361
    Likes Received:
    3
    GPU:
    2x HD7970 - EK Waterblock
    NvLink = CAPI from IBM ( both Nvidia and IBM have developp it, even if they call it differently ), this is basically the Hypertransport Link from AMD or QPI for Intel.. Anyway AMD have a different system for interconnect the systems ( with Supermicro professional system ).

    It can be used anyway for do the connection GPU to GPU, but i dont think it will be used this way for SLI ( too costly and useless )

    This technology effectively dont look to be made for desktop GPU, this is for supercomputer or HPC professional system. you have a cable "mezzanine" who rely the motherboard of one system to the other motherboard for transfer faster the information between a cpu to cpu, or gpu to gpu in the whole system ...

    I dont know what to think about it, comparable system is allready existing and if IBM was allready developp it, this mean all other actors of supercomputers, HPC are doing the same if it is not allready the case.
     
    Last edited: Mar 26, 2014
  10. nicugoalkeper

    nicugoalkeper Master Guru

    Messages:
    866
    Likes Received:
    10
    GPU:
    ASUS GTX 1060 DUAL OC 6GB
    Nice one but still this is NOT FOR DESKTOP PC and NvLink = CAPI is a little to much i think NvLink is way better !
     

  11. Lane

    Lane Ancient Guru

    Messages:
    6,361
    Likes Received:
    3
    GPU:
    2x HD7970 - EK Waterblock
    Nvidia and IBM have team up for develop it as mentionned during the conference by Nvidia, IBM will still call it CAPi ( even if in 2016 this will be CAPi "2.0" or who know what version it will be )...
     
    Last edited: Mar 26, 2014

Share This Page