AlgoFoxGet AlgoFox App

CoursesBuilding Scalable Services

Fan-out and Search

What a write costs when it must reach many readers, the inverted index that makes text searchable, keeping that index in step with the database, and the ranking that trades relevance for freshness.

Lessons
6
Exercises
35
Minutes
39

Start Fan-out by the Numbers

  1. 1

    Fan-out by the Numbers

    After this lesson you can put a number on fan-out on write and fan-out on read for a given follower distribution, and set the threshold where one becomes the other.

    Multiple choiceShort answerTrace
    5 exercises
    6 min
  2. 2

    The Inverted Index

    After this lesson you can say what an inverted index stores, how a two-word query is answered from it, and why a LIKE query cannot do the same job at scale.

    DebugFill blankMultiple choiceShort answer
    5 exercises
    6 min
  3. 3

    Keeping the Index in Sync

    After this lesson you can feed a search index from the database without a dual write, say how far behind it may be, and rebuild it from scratch without losing writes.

    DebugCode orderMultiple choice
    5 exercises
    6 min
  4. 4

    Relevance Against Freshness

    After this lesson you can say what a relevance score rewards, when recency should outrank it, and how to compose the two so that a query for news and a query for a manual get the right order.

    DebugMultiple choiceShort answer
    5 exercises
    6 min
  5. 5

    Checkpoint: Fan-out and SearchCheckpoint

    Fan-out, indexes, sync and ranking for a system you have not seen before.

    DebugMultiple choiceShort answer
    7 exercises
    6 min
  6. 6

    Boss: Marketplace SearchBoss

    One design carried from the ask to the incident: the question, the estimate, the index and its sync, the query contract, the burst of edits, the sold listings that kept appearing, the revised pipeline, and the tradeoff you have to defend.

    Code orderMultiple choiceShort answerTrace
    8 exercises
    9 min