Prospects of gpu tensor core correlation for the sma and the ngeht

HIGHLIGHTS

SUMMARY

    The correlator cross-correlates the baseband data of each antenna/station pair and outputs visibility data for imaging, while the beamformer aggregates the collecting area of the array for VLBI operations. The input data format of TCC and output data format of katgpucbf F-engines are as follows: Input_TCC [ Block_ID ][ Fengine_ID ][ Pol_ID ][ TimePerBlock_ID ]. Output_Fengin [Channel_ID ][ Time_ID ][ Pol_ID ]. As mentioned previously, the throughput of the GPU_Control module is around 140 Gbps, and the main time is spent on memory copying, so the utilization of PCIe4.0 effective bandwidth is approximately 70%. Due to the bandwidth . . .

     

    Logo ScioWire Beta black

    If you want to have access to all the content you need to log in!

    Thanks :)

    If you don't have an account, you can create one here.

     

Scroll to Top

Add A Knowledge Base Question !

+ = Verify Human or Spambot ?