Guide-yakabatana mibvunzo · Zvakaoma Nhanho

CUDA uye GPU Programming Quiz

Check core CUDA concepts, including kernels, SIMT execution, memory access, shared memory, and higher-level GPU programming tools.

Nzira dzekutungamira dzinoenderanaCuda uye GPU Chirongwa
Mubvunzo 1 ye 8

MuCUDA terminology, chii chinonzi 'kernel'?

Tarisa mubvunzo wese8 mibvunzo

Mhinduro uye tsananguro zvinoonekwa paunenge uchitora bvunzo.

  1. MuCUDA terminology, chii chinonzi 'kernel'?
    • Basa rinomhanya richienderana netambo zhinji dzeGPU
    • Iko musimboti weiyo inoshanda sisitimu
    • Chimiro chendangariro chip
    • Iyo GPU yekutonhodza system
  2. Chii chinonzi SIMT execution modhi inorevei?
    • Slow Input, Mazhinji Madhizaini
    • Murairo Mumwechete, Multiple Threads anomhanyisa iwo kuraira kwakafanana pane akasiyana data
    • Mufananidzo Umwechete, Multiple Textures
    • Sequential Mirayiridzo, Manual Nguva
  3. Nei PyTorch uye TensorFlow isingawanzoda vashandisi kunyora mbishi CUDA?
    • MaGPU haakwanise kumhanyisa kodhi yemushandisi
    • CUDA haisi pamutemo kushandisa zvakananga
    • Vanodaidza akakwenenzverwa CUDA maraibhurari senge cuDNN uye cuBLAS mukati
    • Vanomhanya chete paCPUs
  4. Chii chinonzi 'warp divergence' uye nei ichikuvadza kuita?
    • Kukanganisa kwetiweki kunodonhedza mapaketi
    • Iyo GPU inokotama pasi pekupisa
    • Chipembenene chinobvisa ndangariro
    • Threads mu warp inotora matavi akasiyana uye inofanira kumhanya serially
  5. Nei 'coalesced' memory yekuwana yakakosha muCUDA?
    • shinda dzepadhuze kuverenga kero dziri pedyo dzinoshanda zvakanyanya
    • Iyo encrypts data otomatiki
    • Inosunungura dhisiki nzvimbo
    • Inodzima mafeni eGPU
  6. What is OpenAI Triton in relation to CUDA programming?
    • A CUDA driver that replaces NVIDIA GPUs
    • A Python-like language and compiler for writing efficient GPU kernels
    • A model-serving server that only loads TensorFlow checkpoints
    • A framework that converts GPUs into CPUs
  7. What does a CUDA kernel launch configuration specify?
    • A model tokenizer vocabulary
    • The number of training labels
    • The grid and block dimensions used to organize thread execution
    • The host operating system
  8. Why might a CUDA kernel use shared memory?
    • To make data available across every cluster without communication
    • To replace all registers
    • To store the model permanently on the GPU
    • To let threads in a block reuse data from a faster on-chip memory space