feast-dev · zseta · Jun 12, 2026
@@ -15,6 +15,7 @@ Feast integrates with popular vector databases to store and retrieve embedding v
 * **Elasticsearch**: Scalable vector search capabilities
 * **Postgres with PGVector**: SQL-based vector operations
 * **Qdrant**: Purpose-built vector database integration
+* **ScyllaDB**: Native `vector<float, N>` type with HNSW ANN index, full `retrieve_online_documents_v2` support
 
 These integrations allow you to:
 - Store embeddings as features

@@ -15,6 +15,7 @@ Below are supported vector databases and implemented features:
 | Faiss           | [ ]       | [ ]      | []          | []          |
 | SQLite          | [x]       | [ ]      | [x]         | [x]         |
 | Qdrant          | [x]       | [x]      | []          | []          |
+| ScyllaDB        | [x]       | [x]      | [x]         | [x]         |
 
 *Note: V2 Support means the SDK supports retrieval of features along with vector embeddings from vector similarity search.
 
@@ -30,7 +31,7 @@ Beyond that, we will then have `retrieve_online_documents` and `retrieve_online_
 backwards compatibility and the adopt industry standard naming conventions.
 {% endhint %}
 
-**Note**: Milvus and SQLite implement the v2 `retrieve_online_documents_v2` method in the SDK. This will be the longer-term solution so that Data Scientists can easily enable vector similarity search by just flipping a flag.
+**Note**: Milvus, SQLite, and ScyllaDB implement the v2 `retrieve_online_documents_v2` method in the SDK. This will be the longer-term solution so that Data Scientists can easily enable vector similarity search by just flipping a flag.
 
 ## Examples
 

@@ -2,20 +2,15 @@
 
 ## Description
 
-ScyllaDB is a low-latency and high-performance Cassandra-compatible (uses CQL) database. You can use the existing Cassandra connector to use ScyllaDB as an online store in Feast.
-
-The [ScyllaDB](https://www.scylladb.com/) online store provides support for materializing feature values into a ScyllaDB or [ScyllaDB Cloud](https://www.scylladb.com/product/scylla-cloud/) cluster for serving online features real-time.
+[ScyllaDB](https://www.scylladb.com/) is a distributed real-time NoSQL database with vector search support.
+This integration uses the native **`scylla-driver`** Python driver for optimised performance and supports materializing feature values into a [ScyllaDB Cloud](https://www.scylladb.com/product/scylla-cloud/) cluster for real-time online feature serving.
 
 ## Getting started
 
-Install Feast with Cassandra support:
-```bash
-pip install "feast[cassandra]"
-```
+Install Feast with the `scylladb` extra, which pulls in `scylla-driver` automatically:
 
-Create a new Feast project:
 ```bash
-feast init REPO_NAME -t cassandra
+pip install feast[scylladb]
 ```
 
 ### Example (ScyllaDB)
@@ -26,7 +21,7 @@ project: scylla_feature_repo
 registry: data/registry.db
 provider: local
 online_store:
-    type: cassandra
+    type: scylladb
     hosts:
         - 172.17.0.2
     keyspace: feast
@@ -43,36 +38,94 @@ project: scylla_feature_repo
 registry: data/registry.db
 provider: local
 online_store:
-    type: cassandra
+    type: scylladb
     hosts:
         - node-0.aws_us_east_1.xxxxxxxx.clusters.scylla.cloud
         - node-1.aws_us_east_1.xxxxxxxx.clusters.scylla.cloud
         - node-2.aws_us_east_1.xxxxxxxx.clusters.scylla.cloud
     keyspace: feast
     username: scylla
-    password: password
+    password: xxxxxx
+    local_dc: AWS_US_EAST_1
 ```
 {% endcode %}
 
-
-The full set of configuration options is available in [CassandraOnlineStoreConfig](https://rtd.feast.dev/en/master/#feast.infra.online_stores.cassandra_online_store.cassandra_online_store.CassandraOnlineStoreConfig).
-For a full explanation of configuration options please look at file
-`sdk/python/feast/infra/online_stores/contrib/cassandra_online_store/README.md`.
+## Configuration options
+
+| Parameter | Type | Default | Description |
+| :--- | :--- | :--- | :--- |
+| `hosts` | list[str] | *(required)* | Contact-point host addresses. |
+| `port` | int | `9042` | CQL port. |
+| `keyspace` | str | `feast_keyspace` | Target ScyllaDB keyspace. |
+| `username` | str | `None` | Auth username. |
+| `password` | str | `None` | Auth password. |
+| `local_dc` | str | `None` | Local datacenter name for DC-aware load balancing. |
+| `request_timeout` | float | `None` | Driver request timeout in seconds. |
+| `read_concurrency` | int | `100` | `concurrency` argument passed to the driver's `execute_concurrent_with_args` for reads. Controls how many CQL statements are in-flight at once. |
+| `write_concurrency` | int | `100` | `concurrency` argument passed to the driver's `execute_concurrent_with_args` for writes. Controls how many CQL statements are in-flight at once. |
+| `vector_similarity_function` | str | `COSINE` | Default similarity function for vector indexes. Supported: `COSINE`, `DOT_PRODUCT`, `EUCLIDEAN`. Can be overridden per-feature via the `similarity_function` Field tag. |
 
 Storage specifications can be found at `docs/specs/online_store_format.md`.
 
+## Vector Search
+
+ScyllaDB Cloud supports approximate nearest-neighbour (ANN) vector search.
+To enable it for a feature view, tag the embedding `Field` with `vector_index=true` and specify the number of dimensions:
+
+{% code title="feature_definitions.py" %}
+```python
+from feast import FeatureView, Field
+from feast.types import Array, Float32, String
+
+documents_fv = FeatureView(
+    name="documents",
+    entities=[item],
+    schema=[
+        Field(name="text", dtype=String),
+        Field(
+            name="embedding",
+            dtype=Array(Float32),
+            tags={
+                "vector_index": "true",
+                "dimensions": "768",
+                "similarity_function": "COSINE",  # COSINE | DOT_PRODUCT | EUCLIDEAN
+            },
+        ),
+    ],
+    online=True,
+    source=push_source,
+)
+```
+{% endcode %}
+
+When `feast apply` runs, the store creates:
+
+- A regular feature table (`{project}_{fv_name}`) for `online_read` / `online_write_batch`.
+- A vector table (`{project}_{fv_name}__{feature}_vec`) with a native `vector<float, N>` column and an HNSW ANN index.
+
+To query the top-k most similar documents:
+
+```python
+result = store.retrieve_online_documents_v2(
+    features=["documents:text", "documents:embedding"],
+    query=[0.1, 0.2, ...],   # your query embedding
+    top_k=10,
+    distance_metric="COSINE",
+)
+```
+
 ## Functionality Matrix
 
 The set of functionality supported by online stores is described in detail [here](overview.md#functionality).
-Below is a matrix indicating which functionality is supported by the Cassandra plugin.
+Below is a matrix indicating which functionality is supported by the ScyllaDB online store.
 
-|                                                           | Cassandra |
+|                                                           | ScyllaDB  |
 | :-------------------------------------------------------- | :-------- |
 | write feature values to the online store                  | yes       |
 | read feature values from the online store                 | yes       |
 | update infrastructure (e.g. tables) in the online store   | yes       |
 | teardown infrastructure (e.g. tables) in the online store | yes       |
-| generate a plan of infrastructure changes                 | yes       |
+| generate a plan of infrastructure changes                 | no        |
 | support for on-demand transforms                          | yes       |
 | readable by Python SDK                                    | yes       |
 | readable by Java                                          | no        |
@@ -89,6 +142,6 @@ To compare this set of functionality against other online stores, please see the
 
 ## Resources
 
-* [Sample application with ScyllaDB](https://feature-store.scylladb.com/stable/)
+* [ScyllaDB Vector Search documentation](https://cloud.docs.scylladb.com/stable/vector-search/)
 * [ScyllaDB website](https://www.scylladb.com/)
 * [ScyllaDB Cloud documentation](https://cloud.docs.scylladb.com/stable/)
diff --git a/pyproject.toml b/pyproject.toml
@@ -57,6 +57,7 @@ azure = [
     "pymssql<2.3.3"
 ]
 cassandra = ["cassandra-driver>=3.24.0,<4"]
+scylladb = ["scylla-driver>=3.28.0,<4"]
 clickhouse = ["clickhouse-connect>=0.7.19"]
 couchbase = ["couchbase==4.3.2", "couchbase-columnar==1.0.0"]
 delta = ["deltalake<1.0.0"]
@@ -158,7 +159,7 @@ test = [
 ]
 
 ci = [
-    "feast[test, aws, azure, cassandra, clickhouse, couchbase, delta, docling, duckdb, elasticsearch, faiss, gcp, ge, go, grpcio, hazelcast, hbase, ibis, image, k8s, mcp, milvus, mlflow, mongodb, mssql, mysql, openlineage, opentelemetry, oracle, spark, trino, postgres, pytorch, qdrant, rag, ray, redis, singlestore, snowflake, sqlite_vec]",
+    "feast[test, aws, azure, cassandra, clickhouse, couchbase, delta, docling, duckdb, elasticsearch, faiss, gcp, ge, go, grpcio, hazelcast, hbase, ibis, image, k8s, mcp, milvus, mlflow, mongodb, mssql, mysql, openlineage, opentelemetry, oracle, scylladb, spark, trino, postgres, pytorch, qdrant, rag, ray, redis, singlestore, snowflake, sqlite_vec]",
     "build",
     "virtualenv==20.23.0",
     "dbt-artifacts-parser",