Original practice scenario: A hypothetical memory test produces coordinates of failing cells in a very large address space. Compare representations for membership checks, range queries and repeated test runs.
Read model answer +
Choose the representation from defect density and query patterns rather than assuming a bitmap is always compact. Memory-test coordinates also need a stable mapping to the tested device and layout.
Work through it
Normalize coordinates and reject impossible addresses, retaining device, test configuration and run identities so different layouts cannot be accidentally combined.
Compare a hash set, sorted coordinates and compressed ranges or bitmaps, estimating storage and lookup costs for sparse and clustered failures.
Define duplicate behavior and implement range aggregation with boundary tests, including empty maps, maximum addresses and repeated observations across runs.
Tradeoffs to discuss
Hashing offers convenient point lookup but weak ordered scans; sorting improves ranges and compression while increasing update costs.
Common pitfalls
Repeated failures across runs are evidence over time, not necessarily duplicate entries to discard globally.
Do not calculate a defect rate without the tested address population and coverage.
Original practice scenario: A hypothetical quality team changes a memory-testing procedure. Fewer later failures are observed, but the new process tests a different product mix. Propose an evaluation.
Read model answer +
Match observed outcomes to test coverage, product family and follow-up maturity. Fewer reported failures can reflect a smaller or younger deployed population rather than a more effective test.
Work through it
Define the unit under test, defect categories and downstream observation window, linking test results to later outcomes without confusing retests with new units.
Compare like-for-like cohorts or a controlled assignment, reporting coverage and uncertainty separately for product families and test configurations.
Evaluate false rejection as well as escaped defects, and investigate failures seen only under conditions absent from the revised test.
Tradeoffs to discuss
More comprehensive testing can reduce escapes while increasing test time and false rejects; the evaluation should show both consequences.
Common pitfalls
Do not label units with no completed follow-up as confirmed successes.
A change in downstream reporting or usage conditions can alter observed failure rates independently of the test procedure.
