Adaptable Query Accuracy at Scale
Talk data-structures python redis
Social media platforms count likes and comments across billions of records in milliseconds. The trick: they don’t count exactly. I showed how Bloom Filters and other probabilistic data structures make accurate-enough queries possible at scale, and why optimizing for space leads directly to optimizing for time.
The talk included concrete examples with Redis, demonstrating when exact accuracy becomes too expensive and how to configure the error margin for real workloads.