Combining fault simulation and beam data for cnn error rate estimation on risc-v commercial platforms

HIGHLIGHTS

• What: In this research paper the authors assess the induced by neutrons the basic operations of running RISC-V-based processor (GAP8) and how each operation contributes to the entire the findings indicate that memory errors are the primary contributors to the system`s Furthermore the authors present case study demonstrating how the microbenchmarks can be used to estimate the of an entire By combining data from fault simulation and beam experiments the led to result that closely matches those obtained solely from beam experiments. As CNNs are both computationallyand memory-demanding and GAP8 does not have . . .
  

   

  

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

  Thanks :)

  Username or E-mail:

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