Click Loader in the left sidebar to import data from various sources into Ultipa. Supported data sources include CSV, JSON, JSONL, PostgreSQL, SQL Server, MySQL, BigQuery, Neo4j, and Kafka.

Create a Loader

Click New Loader to create a loader. On the loader configuration page, you can rename it and edit the description by clicking the Settings icon next to the loader name.

Next, you can add tasks to the loader. Click the New Task button and choose the type of data source:

To further configure the task, refer to Task Configurations.

Execute Import

Import a Task

You can click the Import button in the task configuration pane to execute a single task.

Run Loader

Click the Run button above the task list triggers the import of all tasks in the loader, following the specified order and processing mode:

• Tasks are executed from top to bottom. You can reorder them by dragging and dropping the task cards.
• Use the S (Serial) or P (Parallel) tag on each task card to toggle its processing modes.
• Consecutive P tasks, along with the S task immediately preceding the first P task, form a parallel group and are imported concurrently.

Errors may occur when importing nodes and edges concurrently, as edge creation relies on the prior existence of their end nodes. To avoid such issues, consider importing nodes before edges, or ensure that the Create Node If Not Exist option is properly configured and the import Mode is set to Insert, Upsert, or Overwrite, as appropriate.

Import Log

A fully successful import is indicated by a green progress bar. If any errors occur during the task, a portion of the progress bar turns yellow, representing the percentage of failed data rows.

The latest import log is displayed at the top of the task configuration pane, where you can also download the Logs for further inspection.

Create Schedule

You can create a schedule to run the loader automatically. Click Create Schedule and configure the trigger strategy:

• Occurrence: Choose One-Time, or set it repeat Weekly or Monthly.
• Timezone: Select the timezone for the scheduled start time.
• Start Time: Specify when the loader should begin execution.

Click Start Schedule to activate it. You can view the execution history from Execution Logs.

Import/Export Loaders

You can export loaders for backup or sharing purposes, and import them to quickly restore or replicate configurations.

Task Configurations

Each task includes two parts of configurations: general settings and data source–specific settings.

General Settings

CSV, JSON, JSONL

File

Example files:

_id:_id,username:string,brithYear:int32
U001,risotto4cato,1978
U002,jibber-jabber,1989
U003,LondonEYE,1982

_from,_to,timestamp
risotto4cato,jibber-jabber,1634962465
LondonEYE,jibber-jabber,1634962587

[
  {
	"_id": "U001",
	"username": "risotto4cato",
	"brithYear": 1978
  },
  {
	"_id": "U002",
	"username": "jibber-jabber",
	"brithYear": 1989
  },
  {
	"_id": "U003",
	"username": "LondonEYE",
	"brithYear": 1982
  }
]

[
  {
	"from": "risotto4cato",
	"to": "jibber-jabber",
	"timestamp": 1634962465
  },
  {
	"from": "LondonEYE",
	"to": "jibber-jabber",
	"timestamp": 1634962587
  }
]

Mapping

PostgreSQL, SQL Server, MySQL

Source Database

SQL

Write SQL to retrieve data from the database, then click Preview to map the results to the properties of nodes or edges.

Mapping

BigQuery

Source Database

SQL

Write SQL to retrieve data from BigQuery, then click Preview to map the results to the properties of the nodes or edges.

Mapping

Neo4j

Neo4j

Cypher

Write Cypher to retrieve nodes or edges from Neo4j, then click Preview to map the results to Ultipa.

Mapping

Kafka

Kafka

Mapping

The Kafka task (consumer), once started, will continue running to import newly published messages. You need to manually stop the task as needed.