1. Performance & Capacity advanced

    A distributed training job runs at 90% scaling efficiency on 512 GPUs. At 1,024 GPUs it drops to 60%. Walk me through where the time went, what you would measure, and what you would try.

    3 min answer nvidiadistributed trainingcollectivesnccl
  2. Performance & Capacity intermediate

    A search engine holds its index entirely in memory to guarantee predictable latency. What does that buy, what does it cost, and when is it the wrong choice?

    2 min answer typesensemeilisearchin-memorypredictability
  3. Performance & Capacity advanced

    Millions of customers attempt to buy a small number of items at a scheduled instant. Which performance constraint dominates, and why do conventional scaling techniques fail against it?

    2 min answer flash-salecontentionhot-keyinventory
  4. Performance & Capacity intermediate

    Peak trading day is six weeks away and expected to be four times normal traffic. What do you do in those six weeks?

    3 min answer capacityload-testingpeakreadiness
  5. Performance & Capacity intermediate

    Precompute every user's timeline at write time, or assemble it at read time? Explain why the answer for a social feed is neither.

    2 min answer case-studytwitterfan-outpower-law
  6. Performance & Capacity intermediate

    Your system handles 1,000 requests per second today. Marketing says a campaign will bring 10,000 next month. What breaks first, and how do you find out?

    2 min answer capacitybottleneckload-testingscaling
  7. Performance Test Design advanced

    A load test shows p99 latency of 120 ms at 2,000 requests per second. Production at the same rate shows 900 ms. Why?

    2 min answer performanceload-testingmethodology
  8. Performance Test Design advanced

    A team must load test at 10× normal traffic and a scaled-down environment tells them nothing. How is load testing in production done safely, and what does it find that nothing else does?

    2 min answer flipkartload-testingproductionisolation
  9. Performance Test Design intermediate Multiple choice

    A test must sustain 2000 requests per second against a service whose p50 is 120 ms, with 1 second of think time per virtual user between requests. Roughly how many virtual users does the generator need and what breaks first?

    2 min answer load-testinglittles-lawvirtual-usersconcurrency
  10. Performance Test Design advanced

    Alibaba has published that it rehearses Singles' Day with full-link stress testing — load generated against the production environment, using traffic shading and data isolation. Your platform faces a comparable annual peak. What does testing in production buy that a staging load test cannot, and what has to be built first?

    4 min answer alibabaload testingtesting in productionshadow data
  11. Performance Test Design advanced

    What makes a performance test predictive of production behaviour, and what is the most common design flaw?

    2 min answer performance-testingdistributionrealismbreaking-point
  12. Performance vs Cost advanced

    A streaming platform can halve latency by doubling infrastructure cost. How should this trade-off be evaluated?

    2 min answer performancecostbusiness-impactsegmentation