GPU Memory Management uye Fragmentation Quiz

Mhinduro uye tsananguro zvinoonekwa paunenge uchitora bvunzo.

1. Chii chinonzi GPU memory fragmentation?
   • Kana iyo GPU inotsemuka mumuviri
   • Kana iyo modhi ine zvidimbu zvakawandisa
   • Kana maGPU maviri asingabvumirani pane mhedzisiro
   • Kana ndangariro yemahara inopatsanurwa kuita akapararira chunks idiki zvakanyanya kugutsa yakakura contiguous chikumbiro
2. Sei masisitimu akaita sePyTorch achishandisa caching memory allocator?
   • Kuita kuti modhi iwedzere kurongeka
   • Kuti udzivise inononoka yakananga mutyairi kugovera / emahara mafoni nekushandisa zvakare akaunganidzwa memory block
   • To encrypt the weights
   • Kudzikisa huwandu hweGPU hunodiwa kune imwe
3. A CUDA allocation fails despite apparently unused reserved memory. What should a developer investigate?
   • Whether fragmentation is proven by the error alone
   • Whether all live tensors have already been freed
   • Whether inactive split blocks or other allocations explain the failure
   • Whether changing the learning rate releases GPU memory
4. Which PyTorch allocator option can be tested for changing-size workloads that leave unused memory slivers?
   • Expandable segments in the allocator configuration
   • Increasing every tensor dimension
   • Disabling the GPU cooling system
   • Increasing the optimizer learning rate
5. Chii chinoita vLLM's PagedAttention inokwanisa kudzikisa tsvina yendangariro panguva yekufungidzira?
   • Data yekudzidzisa yemuenzaniso
   • Iyo yekutarisisa kiyi-kukosha (KV) cache, uchishandisa yakagadziriswa-saizi mapeji
   • Iyo GPU yekumhanyisa wachi
   • Saizi yemafonti ematanda
6. Ndeipi nzira inoderedza activation memory kubatsira kudzivirira kunze-kwe-memory kukanganisa?
   • Kuwedzera mamwe ma layer
   • Kushandisa yakarebesa yekudzidza-chiyero warmup
   • Gradient checkpointing (recomputation)
   • Kuchengetedza mamwe macheki kudhisiki
7. A live CUDA tensor still has Python references. Will empty_cache release that tensor’s allocation?
   • No; it releases unused cached memory rather than live tensor storage
   • Yes; it deletes every referenced tensor
   • Yes; it moves the tensor automatically to NVMe
   • Only if the tensor contains zero values
8. PyTorch reports reserved memory greater than allocated memory. What distinction explains the gap?
   • Reserved memory measures only CPU RAM
   • Allocated memory includes every process on the GPU
   • The difference directly proves a model accuracy error
   • The allocator can retain cached blocks beyond memory currently used by tensors