quiz bu lëkkale ak tegtal · Diggante Niveau

Jamonoy GPU ak Quiz orkestraasioŋ cluster

Jàngale GPU laaj jumtukaay, gang scheduling ak topologie plasement, boole ci ni preemption ak fair-share politik di amee ci clusters tàggat yuñ bokk.

Yooni njiit yu jëm ciGPU ak orkestraasioŋ cluster
Laaj 1 de 8

Why coordinate resource admission for a fixed-size distributed training worker group?

Xoolaatal laaj bu nekk8 laaj

Sooy def quiz bi, tontu yi ak leeral yi dañuy feeñ.

  1. Why coordinate resource admission for a fixed-size distributed training worker group?
    • To guarantee every process starts at exactly the same nanosecond
    • To guarantee the model converges
    • To remove all worker communication
    • To reduce partial-start situations where workers wait for unavailable peers
  2. How does Kubernetes normally express a device-plugin GPU resource request?
    • As arbitrary fractions of CPU clock cycles
    • As an integer count of advertised device resources
    • As a DNS hostname only
    • As a string of model output tokens
  3. Lan la topologie-aware di jéema gëna baaxal?
    • Tasaare liggéey yi fu sori ci seen biir
    • Wàññi limu jëfandikukat yi
    • Teg ay rang yuy jokkoo ci lëkkaloo yu gaaw yu melni NVLink
    • Tann ki gëna xéewale ci joxe kuuraŋ
  4. Ban yokk-on lañuy gëna jëfandikoo ak Kubernetes ngir waajal ay liggéey ci IA?
    • Wolkaan
    • Apache Kafka
    • Redis
    • Grafana
  5. What must a training application provide if a preempted job should resume its previous progress?
    • A guarantee that preemption saves optimizer state automatically
    • A usable checkpoint and recovery procedure
    • A renamed pod with no saved state
    • A higher learning rate for its first resumed step
  6. Lan mooy njeextalu quotas yu jaar yoon ci cluster buñ bokk?
    • Ngir mëna wax ni benn ekip mooy jël ndam li
    • ngir dindi waajalkat bi
    • Ngir moytu benn ekip di monopolise GPU yépp
    • Ngir gaawlu benn liggéey
  7. A fixed worker group needs eight GPU resources but only four are available. What can group admission accomplish?
    • Keep the group pending until its configured minimum requirements can be met
    • Invent the missing four GPU devices
    • Turn every worker into a CPU process without configuration
    • Guarantee completion even with no admitted workers
  8. Two communicating GPU jobs fit the same resource counts on different nodes. Why consider topology as well?
    • The resource count alone fixes every transfer latency
    • Topology determines the training labels
    • Only node names affect all-reduce outputs
    • Interconnect placement affects communication costs between their workers