🎃 We're participating in Hacktoberfest 2023!
Maintained by deepset

Integration: Elasticsearch Document Store

Use an Elasticsearch database with Haystack


The ElasticsearchDocumentStore is maintained within the core Haystack project. It allows you to use Elasticsearch as data storage for your Haystack pipelines.

For a details on available methods, visit the API Reference


To run an Elasticsearch instance locally, first follow the installation and start up guides.

pip install farm-haystack[elasticsearch]

To install Elasticsearch 7, you can run pip install farm-haystac[elasticsearch7].


Once installed, you can start using your Elasticsearch database with Haystack by initializing it:

from haystack.document_stores import ElasticsearchDocumentStore

document_store = ElasticsearchDocumentStore(host = "localhost",
                                            port = 9200,
                                            embedding_dim = 768)

Writing Documents to ElasticsearchDocumentStore

To write documents to your ElasticsearchDocumentStore, create an indexing pipeline, or use the write_documents() function. For this step, you may make use of the available FileConverters and PreProcessors, as well as other Integrations that might help you fetch data from other resources.

Indexing Pipeline

from haystack import Pipeline
from haystack.document_stores import ElasticsearchDocumentStore
from haystack.nodes import TextConverter, PreProcessor

document_store = ElasticsearchDocumentStore(host = "localhost", port = 9200)
converter = TextConverter()
preprocessor = PreProcessor()

indexing_pipeline = Pipeline()
indexing_pipeline.add_node(component=converter, name="TextConverter", inputs=["File"])
indexing_pipeline.add_node(component=preprocessor, name="PreProcessor", inputs=["TextConverter"])
indexing_pipeline.add_node(component=document_store, name="DocumentStore", inputs=["PreProcessor"])


Using Elasticsearch in a Query Pipeline

Once you have documents in your ElasitsearchDocumentStore, it’s ready to be used in any Haystack pipeline. Such as a Retrieval Augmented Generation (RAG) pipeline. Learn more about Retrievers to make use of vector search within your LLM pipelines.

from haystack import Pipeline
from haystack.document_stores import ElasticsearchDocumentStore
from haystack.nodes import EmbeddingRetriever, PromptNode

document_store = ElasticsearchDocumentStore()
retriever = EmbeddingRetriever(document_store = document_store,
prompt_node = PromptNode(model_name_or_path = "google/flan-t5-xl", default_prompt_template = "deepset/question-answering")

query_pipeline = Pipeline()
query_pipeline.add_node(component=retriever, name="Retriever", inputs=["Query"])
query_pipeline.add_node(component=prompt_node, name="PromptNode", inputs=["Retriever"])

query_pipeline.run(query = "Where is Istanbul?")