> Explore Qdrant's agent skills catalog at https://skills.qdrant.tech/
> Search the documentation at https://skills.qdrant.tech/search?query=your+query+here
> Use this file to discover all available pages: https://qdrant.tech/llms.txt

<div class="date">
  <img class="date-icon" src="/icons/outline/date-blue.svg" alt="Calendar" />  Module 4 
</div>


# Where Design Decisions Live

A vector search system has five layers. The first four go from easiest to hardest to change. Distribution is separate, because its cost depends on which change you make, and [Growing Past One Machine](https://qdrant.tech/course/beginners/module-4/growing-past-one-machine/index.md) covers it.

![The five layers of a vector search system as stacked rows. Four sit on a shaded scale from easiest to hardest to change: Query, holding query embedding, dense and sparse search, fusion, and top-K; Indexing, holding the HNSW graph and payload indexes; Storage, holding quantization and on-disk storage; and Data, holding chunking, the embedding model, and the payload schema. Distribution, holding sharding and replication, sits below in a dashed group because what a change costs there depends on the operation.](/courses/beginners/module-4/layers.png)

**Query** handles each request: embed the query, search dense vectors, sparse vectors, or both, then combine the ranked lists into the top-K results. Module 3 covered this layer. If the query is wrong, change it and run it again. The [Query API](https://qdrant.tech/documentation/search/search/index.md) covers every form a query can take.

**Indexing** contains the structures that make search fast. Qdrant builds two of them: the HNSW graph over your vectors, from Module 2, and a payload index over each field you filter on. A mistake here leaves the results correct and makes them slow, and rebuilding the index fixes it. [Indexing](https://qdrant.tech/documentation/manage-data/indexing/index.md) covers how to configure both.

**Storage** controls whether points live in memory or on disk, and therefore how much memory you need. The two main levers are [quantization](https://qdrant.tech/documentation/manage-data/quantization/index.md), which compresses each vector into fewer bytes, and [on-disk vectors](https://qdrant.tech/documentation/manage-data/storage/index.md#configuring-memmap-storage), which keep them in files instead. Both are collection configuration changes, and on an existing collection both rewrite every vector.

**Data** includes the content and the decisions made before it reaches Qdrant. No layer above can fix a mistake here, so the only fix is ingesting the data again. [Decide Before You Ingest](https://qdrant.tech/course/beginners/module-4/decide-before-you-ingest/index.md) works through all four: the text you embed, the model that embeds it, the chunk size, and the payload fields. [Vectors](https://qdrant.tech/documentation/manage-data/vectors/index.md) and [Payload](https://qdrant.tech/documentation/manage-data/payload/index.md) cover what a point can hold.

**Distribution** spreads a collection across more than one machine, through sharding and replication. [Growing Past One Machine](https://qdrant.tech/course/beginners/module-4/growing-past-one-machine/index.md) covers this layer, and [Distributed Deployment](https://qdrant.tech/documentation/distributed_deployment/index.md) has the mechanics.