Skip to main content

create_volume()

This operation creates a volume in the specified project and region.

Request Syntax​

python
create_volume(
    project_id: str,
    region_id: str,
    volume_name: str,
    volume_type: Optional[str] = None,
    storage_integration_id: Optional[str] = None,
    path: Optional[str] = None,
) -> requests.Response

PARAMETERS:

  • project_id (str) -
    [REQUIRED]
    The ID of the Zilliz Cloud project in which to create the volume.

  • region_id (str) -
    [REQUIRED]
    The ID of the Zilliz Cloud region in which to create the volume.

  • volume_name (str) -
    [REQUIRED]
    The name of the volume to create.

  • volume_type (Optional[str]) -
    Default: None
    The volume type. Supported values are MANAGED and EXTERNAL; the default is MANAGED.

  • storage_integration_id (Optional[str]) -
    Default: None
    The storage integration ID required for an EXTERNAL volume.

  • path (Optional[str]) -
    Default: None
    The storage path for an EXTERNAL volume. When omitted, the storage integration root is used; a supplied path must end with /.

RETURN TYPE:

requests.Response

RETURNS:

HTTP response describing the volume creation request.

EXCEPTIONS:

  • MilvusException
    Raised when the server rejects the request or the RPC fails. Inspect the server error message for exact failure details.

Examples​

The example demonstrates create volume usage.

python
from pymilvus.bulk_writer import VolumeFileManager, VolumeManager

manager = VolumeManager(cloud_endpoint="https://api.cloud.zilliz.com", api_key="YOUR_API_KEY")
manager.create_volume(project_id="proj-xxxx", region_id="aws-us-west-2", volume_name="book-volume", volume_type="EXTERNAL")
manager.describe_volume("book-volume")
manager.list_volumes(project_id="proj-xxxx", volume_type="EXTERNAL")

file_manager = VolumeFileManager(cloud_endpoint="https://api.cloud.zilliz.com", api_key="YOUR_API_KEY", volume_name="book-volume")
file_manager.upload_file_to_volume(source_file_path="./data/books.parquet", target_volume_path="datasets/books/books.parquet", upload_concurrency=4)
Ctrl I