Kedis ku xidhan hagaha · Dhexdhexaad Heerka

Kediska Qaybinta Tusaalooyinka Badan ee GPU

Codso muuqaalada MIG, agabka tusaale ahaan GPU-ka-soo-baxa- xisaabinta, iyo soo-gaadhista aaladda, oo ay ku jirto sababta qalabka la taageeray uu u go'aamiyo xulashooyinka qaybinta.

Wadooyinka haga ee la xidhiidhaQaybinta Tusaalooyinka badan ee GPU
Su'aal 1 ee 8

Xuduud kheyraadkee ayaa kala saara MIG GPU Instances ka dibna waqtiga software-ka ee la qaybinayo?

Horudhac su'aal kasta8 su'aalo

Jawaabaha iyo sharraxaadaha waxay soo baxayaan markaad qaadanayso kediska.

  1. Xuduud kheyraadkee ayaa kala saara MIG GPU Instances ka dibna waqtiga software-ka ee la qaybinayo?
    • Only a different process priority on one shared execution queue
    • A guaranteed absence of every host or device failure
    • A separate network cable for every CUDA kernel
    • Dedicated hardware compute and memory resources for each GPU Instance
  2. Qaab dhismeedka NVIDIA ee ugu horeysay ee MIG lagu soo bandhigay?
    • Volta
    • Ampere (A100)
    • Turing
    • Pascal
  3. Astaamaha A100 40 GB oo leh profiles-yada yaryar ee la taageeray, maxay layout-ku siin kartaa tirada GPU Instances?
    • Exactly sixteen on every supported GPU
    • Exactly two regardless of profile
    • Up to seven
    • An unlimited number with no memory limits
  4. Muuqaalka MIG sida '1g.5gb', muxuu '5gb' u taagan yahay?
    • Xusuusta loo qoondeeyay tusaalahaas
    • Tirada SMS-ka
    • bandwidth PCIe
    • Xawaaraha saacada
  5. Shaqadee ayaa MIG gaar ahaan ku habboon?
    • Hal tabobar oo baaxad leh oo u baahan GPU oo dhan
    • Soo jeedinta kiraysteyaal badan oo u adeegaya daahitaan la saadaalin karo
    • Horudhac u samaynta xogta ku xidhan CPU
    • Kaydinta qabow ee muddada-dheer
  6. Sidee kutlada Kubernetes ay u soo bandhigi kartaa tusaalooyinka MIG boobka?
    • Sida mugga diskka cayriin
    • Kaliya iyada oo loo marayo buug-gacmeedka SSH
    • Sida ilaha jadwalaysan sida 'nvidia.com/mig-1g.5gb'
    • Iyada oo loo marayo diiwaannada DNS
  7. Two Compute Instances belong to the same MIG GPU Instance. Which resources do they share?
    • Their parent GPU Instance’s memory resources
    • Only a remote CPU’s filesystem
    • Separate complete copies of physical GPU memory
    • Every compute slice on unrelated GPUs
  8. A team wants to deploy a named MIG profile on a new GPU model. What should it check first?
    • That every NVIDIA GPU supports identical profiles
    • That the model parameter count is an even integer
    • That the Kubernetes pod name contains the profile text
    • That the GPU supports MIG and that exact profile layout