GPU Scheduling and Cluster Orchestration Quiz

Answers and explanations appear as you take the quiz.

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. What does topology-aware scheduling try to optimize?
   • Spreading jobs as far apart as possible
   • Minimizing the number of users
   • Placing communicating ranks on fast interconnects like NVLink
   • Choosing the cheapest electricity provider
4. Which add-on is commonly used with Kubernetes for batch and gang scheduling of AI jobs?
   • Volcano
   • 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. What is the role of fair-share quotas in a shared cluster?
   • To guarantee one team always wins
   • To disable the scheduler
   • To prevent any single team from monopolizing all GPUs
   • To speed up a single job
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