New NVIDIA Pascal GPUs Accelerate Deep Learning Inference

Discussion in 'HardForum Tech News' started by HardOCP News, Sep 13, 2016.

  1. HardOCP News

    HardOCP News [H] News

    Messages:
    0
    Joined:
    Dec 31, 1969
    NVIDIA today unveiled the latest additions to its Pascal™ architecture-based deep learning platform, with new NVIDIA® Tesla® P4 and P40 GPU accelerators and new software that deliver massive leaps in efficiency and speed to accelerate inferencing production workloads for artificial intelligence services. Modern AI services such as voice-activated assistance, email spam filters, and movie and product recommendation engines are rapidly growing in complexity, requiring up to 10x more compute compared to neural networks from a year ago. Current CPU-based technology isn't capable of delivering real-time responsiveness required for modern AI services, leading to a poor user experience.

    The Tesla P4 and P40 are specifically designed for inferencing, which uses trained deep neural networks to recognize speech, images or text in response to queries from users and devices. Based on the Pascal architecture, these GPUs feature specialized inference instructions based on 8-bit (INT8) operations, delivering 45x faster response than CPUs and a 4x improvement over GPU solutions launched less than a year ago. The Tesla P4 delivers the highest energy efficiency for data centers. It fits in any server with its small form-factor and low-power design, which starts at 50 watts, helping make it 40x more energy efficient than CPUs for inferencing in production workloads. A single server with a single Tesla P4 replaces 13 CPU-only servers for video inferencing workloads, delivering over 8x savings in total cost of ownership, including server and power costs.
     
  2. Shintai

    Shintai [H]ardness Supreme

    Messages:
    5,691
    Joined:
    Jul 1, 2016
    A full GP104 running at just 50W is completely amazing.

    2560SP, 160 TMU, 64 ROPs, 810Mhz base, 1063Mhz boost. 256bit BUS, 8GB GDDR5 6Ghz. Passive cooled in a low profile factor.
     
    Ieldra likes this.
  3. Beaflag VonRathburg

    Beaflag VonRathburg Limp Gawd

    Messages:
    399
    Joined:
    Aug 3, 2015
    I thought the Grid M40 cards were impressive (Basically 4x 750ti on a single card), but these architectural improvements shouldl increase performance further. Did the passthrough issues ever get fixed with the M40?
     
  4. Ieldra

    Ieldra I Promise to RTFM

    Messages:
    3,543
    Joined:
    Mar 28, 2016
    i have to say though, the p40 is a little confusing, what distinguishes it from the quadro p6000 other than gddr5 instead of g5x ?

    if there were some damned modding tools you could be underclocking your gtx 1080s for 50w operation lol