跳到主要内容
版本：v2.6.x

VolumeBulkWriter

Adds connection and local output-path behavior.

Request Syntax​

Python
# include-start zilliz
VolumeBulkWriter(
    schema: CollectionSchema,
    remote_path: str,
    cloud_endpoint: str,
    api_key: str,
    volume_name: str,
    chunk_size: int = 1024 * MB,
    file_type: BulkFileType = BulkFileType.PARQUET,
    config: Optional[dict] = None,
    connect_type: ConnectType = ConnectType.AUTO,
    **kwargs,
)
# include-end

PARAMETERS:

RETURN TYPE:

VolumeBulkWriter

RETURNS:

A writer that stages bulk files locally and uploads committed files to the configured Zilliz Cloud volume.

EXCEPTIONS:

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

Examples​

Python
# include-start zilliz
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)
# include-end