Skip to main content
Version: 14.x

Concepts

Data Source​

A data source in the context of Fast Data pipelines represents a static resource that contains data / records not in movement. Example of static data sources are databases and APIs.

Stream​

A stream in the context of Fast Data pipelines represents a channel where data can be moved from one location to another. During the movement, data can be transformed, filtered or aggregated together with other streams. Example of a stream is a Kafka topic.

It is possible to associate a name to each stream, usually similar to the name of the underlying channel. This name can be customized as needed, although it is usually recommended to follow a naming convention. Using the same format for the different stream may help recognizing group of them within the configuration and better understand their purpose.

For example:

  • streams generated as output of a stream-processor service may be suffixed depending on the processor function, such as .validated, .filtered or .preproc.
  • streams generated by a farm-data service may be suffixed with .aggregated to signal their content was combined by joining different sources
  • .aggregated streams transformed and mapped by a stream-processor service may be suffixed as .product, since their content can represent a data product

Snapshot / Initial Load​

The procedure that reads a whole data source and transform each of its records into change events, producing them onto a stream.

Fast Data Message Format​

Within the context of Fast Data, workloads exchange change event following an ad-hoc format. This format provides different details regarding the modification that occurred to the record. In particular, the message contains

  • an identifier of the record that was modified

  • the type of operation it was applied onto the record, which ca be:

    • READ (r) → the record is extracted from the static data source during a snapshot procedure
    • CREATE (c) → the record was just inserted in the static data source
    • UPDATE (u) → the record was updated on the static data source
    • DELETE (d) → the record was deleted from the static data source

  • the value of the record before the modification occurred (available in UPDATE and DELETE operations)

  • the value of the record after the modification occurred (available in CREATE and UPDATE operations)

  • an optional metadata describing the data source that contains the record and when the modification occurred on it

  • an optional metadata describing when the modification was registered by the processing service

warning

In the context of the Fast Data pipelines configuration or this documentation, when a message is required to be Fast Data compliant, it means that it should respect the constraints described in this page and adopt the appropriate message format depending on the selected streaming platform.

Kafka Message Schema​

When using Kafka as stream messaging platform, the identifier of the record must be set in the message key value, while the remaining details must be set in the message payload.

Here is provided the JSON schema of both message components:

Loading ....
Loading ....
note

The raw JSON schema of the message's payload can also be downloaded from here.

Examples​

In this section are provided examples of change events, each of them representing an operation type.

Snapshot Event​

{
  "key": {
    "_id": "60e1d1b1c7849c71e2e3f4g5"
  },
  "payload": {
    "op": "r",
    "after": {
      "_id": "60e1d1b1c7849c71e2e3f4g5",
      "product_id": "P-101",
      "name": "Laptop Pro",
      "price": 1200.00,
      "stock": 50
    },
    "before": null,
    "source": {
      "db": "inventory_db",
      "collection": "products",
      "ts_ms": 1678886400000
    },
    "ts_ms": 1678886401000,
    "updateDescription": null
  }
}

Create Event​

{
  "key": {
    "_id": "60e1d1b1c7849c71e2e3f4g6"
  },
  "payload": {
    "op": "c",
    "after": {
      "_id": "60e1d1b1c7849c71e2e3f4g6",
      "product_id": "P-102",
      "name": "Wireless Mouse",
      "price": 25.50,
      "stock": 200
    },
    "before": null,
    "source": {
      "db": "inventory_db",
      "collection": "products",
      "ts_ms": 1678886402000
    },
    "ts_ms": 1678886403000,
    "updateDescription": null
  }
}

Update Event​

{
  "key": {
    "_id": "60e1d1b1c7849c71e2e3f4g5"
  },
  "payload": {
    "op": "u",
    "after": {
      "_id": "60e1d1b1c7849c71e2e3f4g5",
      "product_id": "P-101",
      "name": "Laptop Pro",
      "price": 1150.00,
      "stock": 49
    },
    "before": {
      "_id": "60e1d1b1c7849c71e2e3f4g5",
      "product_id": "P-101",
      "name": "Laptop Pro",
      "price": 1200.00,
      "stock": 50
    },
    "source": {
      "db": "inventory_db",
      "collection": "products",
      "ts_ms": 1678886404000
    },
    "ts_ms": 1678886405000,
    "updateDescription": null
  }
}

Delete Event​

{
  "key": {
    "_id": "60e1d1b1c7849c71e2e3f4g6"
  },
  "payload": {
    "op": "d",
    "after": null,
    "before": {
      "_id": "60e1d1b1c7849c71e2e3f4g6",
      "product_id": "P-102",
      "name": "Wireless Mouse",
      "price": 25.50,
      "stock": 200
    },
    "source": {
      "db": "inventory_db",
      "collection": "products",
      "ts_ms": 1678886406000
    },
    "ts_ms": 1678886407000,
    "updateDescription": null
  }
}