Shared data directory (Experimental)

In this tutorial we will describe a shared directory data. Qiskit patterns can produce and consume files in the data directory. The files in the data directory are shared among Qiskit patterns.

QiskitServerless has file_download, file_upload, file_delete and files functions that provide file upload, file download, file delete and list files function.

Limitations:

• only tar and h5 files are supported

• tar or h5file should be saved in /data director during your pattern execution to be visible by .files() method call

• only /data directory is supported, /data/other_folder will not be visible

• The working directory of these functions are /data

⚠ This interface is experimental, therefore it is subjected to breaking changes.

⚠ This provider is set up with default credentials to a test cluster intended to run on your machine. For information on setting up infrastructure on your local machine, check out the guide on local infrastructure setup.

import os
from qiskit_serverless import ServerlessClient, QiskitFunction

serverless = ServerlessClient(
    token=os.environ.get("GATEWAY_TOKEN", "awesome_token"),
    host=os.environ.get("GATEWAY_HOST", "http://localhost:8000"),
)
serverless

<ServerlessProvider: gateway-provider>

Upload a file in the data directory

import tarfile

filename= "uploaded_file.tar"
file= tarfile.open(filename,"w")
file.add("manage_data_directory.ipynb")
file.close()

serverless.file_upload(filename)

'{"message":"/usr/src/app/media/mockuser/uploaded_file.tar"}'

Produce a file in the datadirectory

function = QiskitFunction(
    title="file-producer", entrypoint="produce_files.py", working_dir="./source_files/"
)

serverless.upload(function)
job = serverless.run("file-producer")
job.result()

{'Message': 'my_file.txt archived into my_file.tar'}

List files in the data directory

serverless.files()

['uploaded_file.tar', 'my_file.tar']

Consume the file procuded by file-producer Qiskit Pattern

function = QiskitFunction(
    title="file-consumer", entrypoint="consume_files.py", working_dir="./source_files/"
)

serverless.upload(function)
job = serverless.run("file-consumer")
job.result()

{'Message': "b'Hello!'"}

List files

serverless.files()

['uploaded_file.tar', 'my_file.tar']

Delete files

serverless.file_delete("uploaded_file.tar")

'Requested file was deleted.'