---
title: "Managing resources (Hot, Warm, Cold)"
description: "Learn to manage resources effectively in Weaviate for optimized performance."
url: https://docs.weaviate.io/weaviate/starter-guides/managing-resources
---

# Managing resources (Hot, Warm, Cold)

Weaviate provides flexible resource management features that help you to balance search speeds; search accuracy and recall; and system resource costs.

This guide provides an overview of these topics to help you make allocate resources effectively:

- **[Storage tiers (*hot, warm, cold*)](https://docs.weaviate.io/weaviate/starter-guides/managing-resources#storage-tiers---temperatures)**
- **[Vector index types](https://docs.weaviate.io/weaviate/starter-guides/managing-resources#vector-index-types)**
- **[Vector compression](https://docs.weaviate.io/weaviate/starter-guides/managing-resources#vector-compression)**
- **[Tenant states](https://docs.weaviate.io/weaviate/starter-guides/managing-resources#tenant-states)**
- **[Tips](https://docs.weaviate.io/weaviate/starter-guides/managing-resources#tips)**

> [!TIP]
>
> **Resource management Tips**
>
> - Start with the dynamic [index type](https://docs.weaviate.io/weaviate/starter-guides/managing-resources#vector-index-types) unless you have a reason not to.
>
> - Consider [vector compression](https://docs.weaviate.io/weaviate/starter-guides/managing-resources#vector-compression) techniques if some loss of accuracy is acceptable.
>
>   - This will improve query speeds.
>   - For HNSW indexes, this will reduce memory usage.
>
> - Avoid overprovisioning storage. Especially *hot* storage.
>
>   - Can you use [tenant states](https://docs.weaviate.io/weaviate/starter-guides/managing-resources#tenant-states) to reduce costs?
>   - Can you use [vector compression](https://docs.weaviate.io/weaviate/starter-guides/managing-resources#vector-compression) to reduce memory usage?

## Storage Tiers - Temperatures

![Storage Tiers](https://docs.weaviate.io/assets/images/storage-tiers-42d32dfd416335c0a1ff0ffb257d978b.jpg)

We categorize storage resources using three tiers: [*hot*](https://docs.weaviate.io/weaviate/starter-guides/managing-resources#-hot), [*warm*](https://docs.weaviate.io/weaviate/starter-guides/managing-resources#-warm), and [*cold*](https://docs.weaviate.io/weaviate/starter-guides/managing-resources#-cold). Each tier has different performance characteristics and costs.

| Tier | Vector Index Type | Vector Compression | Tenant State | Storage | Performance | Cost |
| - | - | - | - | - | - | - |
| 🟥 Hot | HNSW | PQ, SQ, BQ | Active | Memory | Fastest | High |
| 🟨 Warm | Flat | BQ | Active | SSD | Slower | Moderate |
| 🟦 Cold | Any | Any | Inactive | Cloud | Resource not available | Low |

### 🟥 Hot

![Storage Tiers - Hot](https://docs.weaviate.io/assets/images/storage-tiers-hot-3e08f3be3955382d81a8f8641b7b542d.jpg)

- Describes memory usage
- Fastest and most expensive
- Primarily driven by [`HNSW`](https://docs.weaviate.io/weaviate/starter-guides/managing-resources/indexing#hnsw-indexes) vector indexes
- Always available (active) for use
- Costs increase rapidly with scale

### 🟨 Warm

![Storage Tiers - Warm](https://docs.weaviate.io/assets/images/storage-tiers-warm-abd285542c887cfb7776cbfbf9132551.jpg)

- Describes data stored on disk (SSD)
- Slower than [hot](https://docs.weaviate.io/weaviate/starter-guides/managing-resources#-hot) tier but less expensive
- Driven by [flat](https://docs.weaviate.io/weaviate/starter-guides/managing-resources/indexing#flat-indexes) vector index, object data, and [inverted indexes](https://docs.weaviate.io/weaviate/starter-guides/managing-resources/indexing#inverted-indexes)
- Always available (active) for use
- Costs increase more slowly than hot tier as data grows

### 🟦 Cold

![Storage Tiers - Cold](https://docs.weaviate.io/assets/images/storage-tiers-cold-426d159a89cb78695d9e12486c1a2549.jpg)

> [!NOTE]
>
> **Offloading: AWS S3 only**
>
> As of Weaviate `v1.26.0`, tenants can only be offloaded to cold storage in AWS S3. Additional storage options may be added in future releases.
>
> To offload a tenant, use the `offload-s3` module.

- Describes data stored in cloud storage
- Slowest and least expensive tier
- Primarily driven by [offloaded tenants](https://docs.weaviate.io/weaviate/starter-guides/managing-resources#tenant-states)
- Resources are not available (inactive) for use
- Requires [reactivation](https://docs.weaviate.io/weaviate/starter-guides/managing-resources#tenant-states) to access

## Resource Management - Key Factors

Effective resource management in Weaviate involves balancing performance, cost, and data accessibility. The key levers to manage resources are:

- **[Vector index types](https://docs.weaviate.io/weaviate/starter-guides/managing-resources#vector-index-types)**: Choose the right index type based on the number of objects and desired performance.
- **[Vector compression](https://docs.weaviate.io/weaviate/starter-guides/managing-resources#vector-compression)**: Use compression techniques to reduce memory usage and improve query performance at the cost of some accuracy.
- **[Tenant states](https://docs.weaviate.io/weaviate/starter-guides/managing-resources#tenant-states)**: Manage tenant states to balance cost and performance.

### Vector index types

The choice of vector index type can have a significant impact on performance and resource usage. Weaviate supports the following index types:

| Index Type | Resource Usage | Performance | Suitable for | Description |
| - | - | - | - | - |
| HNSW | 🟥 Hot | Fast | Any object count | A memory-based, fast index ([read more](https://docs.weaviate.io/weaviate/starter-guides/managing-resources/indexing#hnsw-indexes)) |
| Flat | 🟨 Warm | Medium | <\~10k objects | A disk-based, brute-force index ([read more](https://docs.weaviate.io/weaviate/starter-guides/managing-resources/indexing#flat-indexes)) |
| Dynamic | Depends | Depends | Any object count | Transitions from flat to HNSW index at a specified threshold ([read more](https://docs.weaviate.io/weaviate/starter-guides/managing-resources/indexing#dynamic-indexes)) |
| HFresh | 🟨 Warm | Fast | Large collections | A cluster-based index with memory efficiency for large datasets ([read more](https://docs.weaviate.io/weaviate/starter-guides/managing-resources/indexing#hfresh-indexes)) |

The choice of index type depends on the number of objects and the desired performance. As a rule of thumb, use the following guidelines for a multi-tenant collection:

```mermaid
flowchart LR
    %% Define nodes and connections
    subgraph vectorIndex ["Vector Index Selection"]
        direction LR
        multiTenant{"Is multi-tenancy<br> enabled?"}
        objCount{"Does the collection have<br> more than 100k objects?"}
        stayUnder{"Is the collection likely to stay<br> at under 100k objects?"}

        multiTenant -->|No| objCount
        multiTenant -->|Yes| useDynamic["Use Dynamic"]
        objCount -->|Yes| useHNSW["Use HNSW"]
        objCount -->|No| stayUnder
        stayUnder -->|Yes| useFlat["Use Flat"]
        stayUnder -->|"No/Unsure"| useDynamic["Use Dynamic"]
    end

    %% Style nodes
    style multiTenant fill:#ffffff,stroke:#B9C8DF,color:#130C49
    style objCount fill:#ffffff,stroke:#B9C8DF,color:#130C49
    style stayUnder fill:#ffffff,stroke:#B9C8DF,color:#130C49

    %% Terminal nodes
    style useHNSW fill:#ffffff,stroke:#61BD73,color:#130C49
    style useFlat fill:#ffffff,stroke:#61BD73,color:#130C49
    style useDynamic fill:#ffffff,stroke:#61BD73,color:#130C49

    %% Style subgraph
    style vectorIndex fill:#ffffff,stroke:#130C49,stroke-width:2px,color:#130C49
```

If you are unsure which index type to use, the dynamic index type is a good starting point, as it automatically transitions from a flat to an HNSW index based on the number of objects.

- [Starter guide: indexes](https://docs.weaviate.io/weaviate/starter-guides/managing-resources/indexing)
- [How-to: Set the vector index type](https://docs.weaviate.io/weaviate/manage-collections/vector-config#set-vector-index-type)

### Vector compression

Vector compression techniques reduce the size of vectors by quantizing them into a smaller representation.

This can have the impact of reducing memory usage, or improving performance by reducing the amount of data that needs to be read from disk. The trade-off is that the resulting search quality may be lower.

Weaviate supports the following vector compression methods:

| Compression Method | Index Type | Requires Training | Description |
| - | - | - | - |
| Product Quantization (PQ) | HNSW | Yes | Each vector becomes an array of integer-based centroids ([read more](https://docs.weaviate.io/weaviate/concepts/vector-quantization#product-quantization)) |
| Binary Quantization (BQ) | HNSW, Flat | No | Each vector dimension becomes a bit ([read more](https://docs.weaviate.io/weaviate/concepts/vector-quantization#binary-quantization)) |
| Scalar Quantization (SQ) | HNSW | Yes | Each vector dimension becomes an integer ([read more](https://docs.weaviate.io/weaviate/concepts/vector-quantization#scalar-quantization)) |
| Rotational Quantization (RQ) | All index types | No | Each vector is rotated then quantized to an integer ([read more](https://docs.weaviate.io/weaviate/concepts/vector-quantization#rotational-quantization)) |

As a starting point, use the following guidelines for selecting a compression method:

```mermaid
flowchart LR
    %% Define nodes and connections
    subgraph compression ["Compression Strategy"]
        direction LR
        startIndexType{"What is your<br> vector index type?"}
        sample{"Do you have a<br> representative sample<br> of your final dataset?"}
        tunable{"Do you want<br> tunable<br> compression?"}
        bqcompat{"Is your vectorizer<br> model compatible<br> with BQ?"}

        startIndexType -->|HNSW| sample
        startIndexType -->|Flat| bqcompat
        sample -->|Yes| tunable
        tunable -->|Yes| usePQ["Use PQ"]
        tunable -->|No| useSQ["Use SQ"]
        sample -->|No| useBQ["Use BQ"]
        bqcompat -->|Yes| useBQ["Use BQ"]
        bqcompat -->|No| noCompress["Do not use<br> compression"]
    end

    %% Style nodes
    style startIndexType fill:#ffffff,stroke:#B9C8DF,color:#130C49
    style sample fill:#ffffff,stroke:#B9C8DF,color:#130C49
    style tunable fill:#ffffff,stroke:#B9C8DF,color:#130C49
    style bqcompat fill:#ffffff,stroke:#B9C8DF,color:#130C49

    %% Terminal nodes
    style usePQ fill:#ffffff,stroke:#61BD73,color:#130C49
    style useSQ fill:#ffffff,stroke:#61BD73,color:#130C49
    style useBQ fill:#ffffff,stroke:#61BD73,color:#130C49
    style noCompress fill:#ffffff,stroke:#61BD73,color:#130C49

    %% Style subgraph
    style compression fill:#ffffff,stroke:#7AD6EB,stroke-width:2px,color:#130C49
```

If you are unsure which index type to use, scalar quantization is a good starting point, provided that you have a representative sample of your likely final dataset.

- [Starter guide: Vector compression](https://docs.weaviate.io/weaviate/starter-guides/managing-resources/compression)
- [How-to: Configure vector compression](https://docs.weaviate.io/weaviate/configuration/compression)

### Tenant states

Multi-tenant collections enable you to efficiently manage isolated subsets of data. Each tenant share the same schema and configuration.

Weaviate supports the following tenant states:

| Tenant state | CRUD & Queries | Vector Index | Inverted Index | Object Data | Time to Activate | Description |
| - | - | - | - | - | - | - |
| Active (default) | **Yes** | Hot/Warm | Warm | Warm | None | Tenant is available for use |
| Inactive | **No** | Warm | Warm | Warm | Fast | Tenant is locally stored but not available for use |
| Offloaded | **No** | Cold | Cold | Cold | Slow | Tenant is stored in cloud storage and not available for use |

*Hot* tenants can be deactivated to *warm* storage to reduce memory usage, and any tenant can be offloaded to *cold* storage to reduce memory and disk usage. Conversely, any tenant can be reactivated when needed.

```mermaid
flowchart LR
    %% Define nodes and connections
    subgraph tenantData ["Tenant Data Availability"]
        direction LR
        needNow{"Does the tenant data<br> need to be<br> available now?"}
        howQuick{"When it is needed,<br> how quickly does it<br> need to be available?"}

        needNow -->|Yes| active["Active"]
        needNow -->|No| howQuick
        howQuick -->|Quickly| inactive["Inactive"]
        howQuick -->|"Latency is acceptable"| offloaded["Offloaded"]
    end

    %% Style nodes
    style needNow fill:#ffffff,stroke:#B9C8DF,color:#130C49
    style howQuick fill:#ffffff,stroke:#B9C8DF,color:#130C49

    %% Terminal nodes
    style active fill:#ffffff,stroke:#61BD73,color:#130C49
    style inactive fill:#ffffff,stroke:#61BD73,color:#130C49
    style offloaded fill:#ffffff,stroke:#61BD73,color:#130C49

    %% Style subgraph
    style tenantData fill:#ffffff,stroke:#61BD73,stroke-width:2px,color:#130C49
```

Consider a strategy of deactivating tenants that are not frequently accessed, and offloading tenants that are rarely accessed.

- [Starter guide: tenant states](https://docs.weaviate.io/weaviate/starter-guides/managing-resources/tenant-states)
- [How-to: Configure tenant offloading](https://docs.weaviate.io/deploy/configuration/tenant-offloading)
- [How-to: Manage tenant states](https://docs.weaviate.io/weaviate/manage-collections/tenant-states)

## Tips

### Best Practices

- Start with the dynamic [index type](https://docs.weaviate.io/weaviate/starter-guides/managing-resources#vector-index-types) for new collections. This is particularly useful for multi-tenant collections, as it allows each tenant to use the most appropriate index type.
- Use [vector compression](https://docs.weaviate.io/weaviate/starter-guides/managing-resources#vector-compression) techniques to optimize storage and query performance, especially for large collections or tenants.
- Conduct thorough testing when changing index types or compression methods to ensure performance meets your requirements.

### Common Pitfalls

- Overprovisioning hot storage: Keeping all data in hot storage can lead to unnecessary costs. Regularly assess what data truly needs the fastest access.
- Neglecting to plan for growth: Not anticipating data growth can lead to performance issues. Always design your resource management strategy with scalability in mind.
- Improper tenant management: In multi-tenant scenarios, forgetting to [offload inactive tenants](https://docs.weaviate.io/weaviate/starter-guides/managing-resources#tenant-states) can lead to resource waste. Implement automated processes to manage tenant states based on usage patterns.
- Mismatch between quantization techniques, model and data: When using compression technique, ensure that the quantization technique is compatible with the model (e.g. BQ) and that the data is sufficient and representative for training (e.g. PQ, SQ).

## Related pages

- [Starter guide: Compression](https://docs.weaviate.io/weaviate/starter-guides/managing-resources/compression)
- [Starter guide: Indexing](https://docs.weaviate.io/weaviate/starter-guides/managing-resources/indexing)
- [Starter guide: Tenant states](https://docs.weaviate.io/weaviate/starter-guides/managing-resources/tenant-states)
- [Concepts: Vector Index](https://docs.weaviate.io/weaviate/concepts/vector-index)
- [Concepts: Vector Quantization](https://docs.weaviate.io/weaviate/concepts/vector-quantization)
- [Concepts: Multi-Tenancy](https://docs.weaviate.io/weaviate/concepts/data#multi-tenancy)
- [How-to: Set the vector index type](https://docs.weaviate.io/weaviate/manage-collections/vector-config#set-vector-index-type)
- [How-to: Configure vector compression](https://docs.weaviate.io/weaviate/configuration/compression)
- [How-to: Perform multi-tenancy operations](https://docs.weaviate.io/weaviate/manage-collections/multi-tenancy)
- [How-to: Manage tenant states](https://docs.weaviate.io/weaviate/manage-collections/tenant-states)
