Skip to main content
Version: User Guides (Cloud)

FAQ: Migration

This topic lists the possible issues that you may encounter while you migrate data on Zilliz Cloud and the corresponding solution.



How to migrate data from Elasticsearch to Zilliz Cloud?

You can follow the steps below to import data from Elasticsearch to Zilliz Cloud.

  1. Connect to ElasticSearch.

    print(fmt.format("start connecting to ElasticSearch"))
    es = Elasticsearch(hosts=ELASTICSEARCH_HOST)
  2. Use the Elasticsearch Scroll API to retrieve datasets from Elasticsearch and import them into Zilliz Cloud. The following example imports the Elasticsearch data into a collection named hello_milvus in Zilliz Cloud.

    print(fmt.format("Start inserting entities"))
    rng = np.random.default_rng(seed=19530)
    resp =, query={"match_all": {}}, scroll="1m", size=500)
    scroll_id = resp["_scroll_id"]

    query = {
    "scroll": "1m",
    "scroll_id": scroll_id,

    # Start the primary key from 0
    pk = 0
    while len(resp["hits"]["hits"]):
    l = len(resp["hits"]["hits"])
    entities = [[str(i + pk) for i in range(l)], rng.random(l).tolist()]
    embeddings = []
    for item in resp['hits']['hits']:
    insert_result = hello_milvus.insert(entities)
    pk += l
    resp = es.scroll(scroll_id=scroll_id, scroll="1m")
  3. Check the number of entities you have imported.

    print(f"Number of entities in Milvus: {hello_milvus.num_entities}")

Alternatively, you can also import data from Elasticsearch on the Zilliz Cloud web console. For more information, see Migrate from Elasticsearch .

Can I migrate data from a serverless cluster to a dedicated cluster?

Yes. Zilliz Cloud supports migrating data from a serverless cluster to a dedicated cluster. For more information about how to migrate data, please refer to Migrate Between Clusters.