Query Optimization (filters, indexes, ordering)

Query optimization restructures work—push filters early, index joins, batch network round trips. Same spirit in contests: answer many range questions with one sweep instead of independent scans.

Why this shows up in the real world

SQL planners reorder joins for cardinality. GraphQL batching reduces HTTP chatter.

Core idea (explained for students)

Reorder operations by selectivity: cheap filters first. Use auxiliary indexes (sorted list + bisect, hash buckets) to avoid full scans.

Try this in Python

def filter_then_map(nums: list[int], pred, fn):
    return [fn(x) for x in nums if pred(x)]


print(filter_then_map([1, 2, 3, 4, 5], lambda x: x % 2 == 0, lambda x: x * x))

Common mistakes

• Optimizing average case while worst case still timeouts hidden tests.
• Cache poisoning when query results depend on mutable global state.

Key takeaways

• Capture representative workloads before tuning.
• Prefer declarative filters in code for readability, then tighten hotspots.