Supercomputing 2018: EPYC, Immersion and Quantum Computing

About the author
Paul Alcorn

Paul Alcorn is a Contributing Editor for Tom's Hardware US. He writes news and reviews on CPUs, storage and enterprise hardware.

Read more
Create a new thread in the Photo reports comments forum about this subject
4 comments
Comment from the forums
    Your comment
  • redgarl
    Intel MCM is not AMD Infinity Fabric Chiplet design. They are really different. One is real glue and the other is a modular approach. As of now, Intel is only able to plug 2 monolithic chips together like they did with the core 2 duo.
  • redgarl
    "Intel's top-end Platinum -AP models, which will reportedly stretch up to 350W, will require liquid cooling for stock operation."

    Who the hell at Intel thought this would be a good idea?
  • attb072
    there is a market, demand for this. AND WHY THE HELL SHOULD INTEL BE SHOWN UP AS NOT BEING HIGH-HIGH END NON COMPETITIVE. THAT AIN'T HIP.
  • bit_user
    Anonymous said:
    It is thought that over 1,024 TPU v3 processors will come together to form a single pod. Google claims a full pod can deliver well over one hundred petaflops of performance.

    Google isn't sharing many specifics of the v3 models, though we do know they can deliver 420 teraflops of performance and come with 128GB of HBM.

    It sounds like the 420 TFLOPS spec is for a 4-TPU board. Then, a single TPU should be good for about 105 TFLOPS, putting it just below a Nvidia V100.

    Hmmm...