Skip to content

Import

Import is a guided process for importing data from a number of different sources, including cloud and relational data services.

Available data sources to import data from.
Available data sources to import data from.

Set up

  1. Select a data source; a guided walkthrough for Object Storage, Kafka or SQL database imports via the UI are available to try.

  2. Complete the workflow, then Open Pipeline.

  3. Changes can be made in the pipeline; for example, for Kafka import it is necessary to include a Map function.

  4. Save the pipeline (giving it a name in the process).

  5. Deploy the pipeline to an active database, or deploy a database and associated pipeline(s). A successfully deployed database shows a green circle with a tick inside it. The status of a pipeline is available under Running Pipelines in the Overview page; a successfully deployed pipeline lists as Running or Finished under Status and has a green tick or grey circle respectively next to it.

    Guided Walkthrough

    If using guided walkthrough examples, create and deploy an insights-demo database before deploying the pipeline.

  6. Open a Query tab to query the data. Opening the SQL tab is the quickest way to get data:

    SELECT * FROM tablename
    

Advanced Settings

Each pipeline created by the Import wizard will have an option for Advanced pipeline settings. It should be noted, no additional code is required for the standard import options described below. The Advanced settings include Global Code which is executed before a pipeline is started; this code may be necessary for Stream Processor hooks, global variables and named functions.

Schema Column Parsing

Parsing is done automatically as part of the import process. If defining a manual parse, ensure parsing is enabled for all time, timestamp, and string fields unless your input is IPC or RT.

Parsing is done automatically, but it should be done for all time, timestamp and string fields if not automatic.
Parsing is done automatically, but it should be done for all time, timestamp and string fields if not automatic.

Kafka

Connect to an Apache Kafka distributed event streaming platform.

  1. Define the Reader node with the Kafka broker details, port information and topic name.

    Sample Kafka Reader Node

    • Broker: kafka:9092 (note:not a working example)
    • Topic: trades
    • Offset: Start
    • Use TLS: No
  2. Event data on Kafka is of JSON type, so use a JSON decoder node to transform the data to a kdb+ friendly format. Data will be converted to a kdb+ dictionary. There is no need to change the default setting for this.

    JSON decode

    • Decode Each: false
  3. Apply a Schema with the Transform node; this ensures data is of the correct datatype in the destination database. A schema can be loaded from a database; use the Build a Database wizard available from the Overview page to create a database and schema if required. Parse Strings can be left as Auto; but ensure any time, timestamp and string data columns is set to On or Auto.

    Sample Kafka transform for trades

    • Date Format: Any
    column name column type parse strings
    datetime Timestamp Auto
    ticker Symbol Auto
    price Float Auto
    size Integer Auto

    I want to learn more about schemas.

  4. Write the data to a table.

    item description
    Database Select an database from the dropdown to assign the data, or create a database by clicking Add Database icon, before returning to the import wizard to add the database.
    Table Select a table from the selected database to write the data to; this is typically the schema table from the previous step. If not completed during the schema step, create a database with a schema matching the data.
    Write Direct to HDB Unchecked
    Deduplicate Stream Can remain checked.

    I want to learn more about databases.

  5. Open the Pipeline view

  6. Include a Map function to convert the data from a kdb+ dictionary to a kdb+ table. This Map function should be inserted between the JSON Decoder of Step 2 and the Transform Schema of Step 3 in the pipeline view; see image below:

    A completed Kafka pipeline.
    A completed Kafka pipeline.

    Map function

    {[data]
        enlist data
    }
    
  7. Save the Pipeline

  8. Deploy the database and Kafka pipeline.

  9. Query the data. Open the SQL tab. Use the topic name as the Output variable. This will be the name of the table to query when getting data; for example, SQL query:

    SQL query

    SELECT * FROM topicname
    
  10. Get Data

The Broker field can be the Kafka broker pod name when the Kafka broker is installed in the same Kubernetes namespace as the KX Insights Platform. If the Kafka cluster has been installed into a different Kubernetes namespace, then the field expects the full service name of the Kafka broker - i.e. kafka:9092, where default is the namespace that the Kafka cluster is in. The full-service name is set in the KX_KAFKA_BROKERS environment variable property in the Kafka Producer Kubernetes installation instructions.

See the guided walkthrough Kafka import example.

Kdb Expression

Import data directly using a kdb+ expression.

  1. The Kdb Expression Reader starts with a code editor; add the following kdb+ data generator in the node editor:

    kdb+ query

    n:2000;
    ([] date:n?(reverse .z.d-1+til 10);
        instance:n?`inst1`inst2`inst3`inst4;
        sym:n?`USD`EUR`GBP`JPY;
        cnt:n?10)
    
  2. Configure the Schema to use. This can be done manually within the node, or preferably, loaded from a database.

    Data Format
    Any
    Schema
    Use the defined schema for the kdb expression import, created manually or imported from an existing schema. Parse Strings can be set to Auto for all columns; but ensure any time, timestamp and string data columns is set to On or Auto.
    column name column type parse strings
    date Timestamp Auto
    instance Symbol Auto
    symbol Symbol Auto
    cnt Integer Auto

    I want to learn more about schemas.

    A schema can also be loaded from a database by clicking Schema create button. The data parse is defined in the schema tab of the database.

    An expression pipeline schema.
    An expression pipeline schema.

    I want to learn more about schemas.

  3. Configure the Writer.

    item description
    Database Select an database from the dropdown to assign the data, or create a database by clicking Add Database icon before returning to the wizard to add the database.
    Table Select a table from the selected database to write the data to; this is typically the schema table from the previous step. If not completed during the schema step, create a database with a schema matching the data.
    Write Direct to HDB Unchecked
    Deduplicate Stream Can remain checked.

    I want to learn more about databases.

  4. Open the Pipeline

  5. Save the Pipeline

    Saving an expression pipeline.
    Saving an expression pipeline.

  6. Add the pipeline to the database. Free Trials users can add the pipeline to their insights-demo database. Save the database with the pipeline.

    Expression pipeline associated with the insights-demo database.
    Expression pipeline associated with the insights-demo database.

  7. Deploy the Data database and associated Expression pipeline.

  8. Query the data. Open the SQL tab. Repeat the schema table name as the Output variable. This will be the name of the table to query when getting data; for example:

    SQL query

    SELECT * FROM tablename
    
  9. Get Data

SQL Databases

Applicable to all supported SQL databases, including PostgreSQL and Microsoft SQL Server.

  1. Define the Reader node, including any required authentication, alongside server and port details.

    Sample Postgres Reader Node

    • Server: postgresql
    • Port: 5432
    • Database: testdb
    • Username: testdbuser
    • Password: testdbuserpassword
    • query: select * from cars

    Where Server is postgresql.default.svc, the default is the namespace, e.g. db will be postgresql.db.svc

    case sensitive

    Node properties and queries are case sensitive and should be lower case.

  2. Define the Schema. Parse Strings can be set to Auto for all columns; but ensure any time, timestamp and string data columns is set to On or Auto.

    Sample Postgres Configure Schema

    • Data Format: Any
    column name column type parse string
    Year Date Auto
    Name Symbol Auto
    Miles_per_Gallon Float Auto
    Cylinders Float Auto
    Displacement Float Auto
    Horsepower Float Auto
    Acceleration Float Auto
    Origin Symbol Auto

    A schema for a sample pipeline.
    A schema for a sample pipeline.

    I want to learn more about schemas.

  3. Configure the Writer

    item description
    Database Select an database from the dropdown to assign the data, or create a database by clicking Add Database icon, before returning to the import wizard to add the database.
    Table Select a table from the selected database to write the data to; this is typically the schema table from the previous step. If not completed during the schema step, create a database with a schema matching the data.
    Write Direct to HDB Unchecked
    Deduplicate Stream Can remain checked.

    I want to learn more about databases.

  4. Open Pipeline

  5. Save the Pipeline

  6. Deploy the database and Postgres pipeline.

  7. Query the data. Open the SQL tab. Define an Output variable, e.g. "cars", this will be the name of the table to query when using kdb+/q or python in the scratchpad.

    For querying in the SQL tab:

    SQL query

    SELECT * FROM cars
    
  8. Get Data

See the guided walkthrough SQL database import example.

Cloud Storage Services

Organizations require data services which balance performance, control, security and compliance - while reducing the cost of storing and managing such vast amounts of data. Limitless scale is the reason object storage is the primary storage of cloud service providers, Amazon, Google and Microsoft. kdb Insights allows users take advantage of object storage and an example is available as part of a guided walkthrough

  1. Define the Reader node with the S3 path details for how the file will be read and optionally, the Kubernetes secret for authentication. Default values can be used for all but Path and Region.

    Sample Amazon 3 Reader Node

    • Path: s3://bucket/pricedata.csv
    • Region: us-east-1
  2. Define a CSV decoder node. Your imported data source requires a timestamp column to parse the data.

    Typical CSV decoder Node

    • Delimiter: ","
    • Header: First Row
    • Encoding Format: UTF8

    Other required properties can be left at default values.

  3. Define the Transform Schema application

    A sample Schema for a time column, category column and two value columns

    • Data Format: Any
    • Schema: Load a schema from a database by clicking schema add button; create a database and schema if needed, or manually add the columns and column type for each of the source data columns in the Transform Node. Parse Strings can be set to Auto for all columns; but ensure any time, timestamp and string data columns is set to On or Auto.
    column name column type parse strings
    time Timestamp Auto
    sym Symbol Auto
    price Float Auto
    volume Integer Auto

    I want to learn more about schemas.

  4. Define the table name to upsert the data to. Select the database and table to write the data too from the dropdown menu; this will be the same database and table from the schema creation step.

    Writer Node

    • Database: databasename
    • Table: pricedata
    • Write Direct to HDB: unchecked
    • Deduplicate stream: checked

    I want to learn more about databases

  5. Open the Pipeline

  6. Save the Pipeline

  7. Deploy the database with associated Amazon S3 pipeline.

  8. Query the data. Open the SQL tab. Define the Output variable as "pricedata".

    SQL query

    SELECT * FROM pricedata
    
  9. Get Data

Note that a Kubernetes secret is required if authenticating with Amazon S3. See the section on Amazon S3 Authentication for more details.

I want to learn more about Amazon S3 Authentication.

  1. Define the Reader node with GCS details. Default values can be used for all but Path and Project ID. Project ID may not always have a customizable name, so its name will be contingent on what's assigned when creating your GCS account.

    Sample GCS Reader Node

    • Path: gs://mybucket/pricedata.csv
    • Project ID: ascendant-epoch-999999

    Other required properties can be left at default values.

  2. Define a CSV decoder node. Your imported data source requires a timestamp column to parse the data.

    Typical CSV decoder Node

    • Delimiter: ","
    • Header: First Row
    • Encoding Format: UTF8
  3. Define the Transform Schema application

    A sample Schema for a time column, category column and two value columns

    • Data Format: Any
    • Schema: Load a schema from a database by clicking schema add icon; create a database and schema if needed, or manually add the columns and column type for each of the source data columns in the Transform Node. Parse Strings can be set to Auto for all columns; but ensure any time, timestamp and string data columns is set to On or Auto.
    column name column type parse strings
    time Timestamp Auto
    sym Symbol Auto
    price Float Auto
    volume Integer Auto

    I want to learn more about schemas.

  4. Define the table name to upsert the data to. Select the database and table to write the data to from the dropdown menu; this will typically be the same database and table from the schema creation step.

    Writer Node

    • Database: databasename
    • Table: pricedata
    • Write Direct to HDB: unchecked
    • Deduplicate stream: checked

    I want to learn more about databases

  5. Open the Pipeline

  6. Save the Pipeline

  7. Deploy the database and associated GCP pipeline.

  8. Query the data. Open the SQL tab. Define the Output variable as "pricedata".

    SQL query

    SELECT * FROM pricedata
    
  9. Get Data

I want to learn more about Google Cloud Storage Authentication

  1. Define the ACP Reader node with ACS details. Default values can be used for all but Path and Project ID. A customizable account name is supported for ACS.

    Sample ACP Reader Node

    • Path: ms://mybucket/pricedata.csv
    • Account: myaccountname

    Other required properties can be left at default values.

  2. Define a CSV decoder node. Your imported data source requires a timestamp column to parse the data.

    Typical CSV decoder Node.

    • Delimiter: ","
    • Header: First Row
    • Encoding Format: UTF8
  3. Define the Transform Schema application

    A sample Schema for a time column, category column and two value columns

    • Data Format: Any
    • Schema: Load a schema from a database by clicking schema add icon; create a database and schema if needed, or manually add the columns and column type for each of the source data columns in the Transform Node. Parse Strings can be set to Auto for all columns; but ensure any time, timestamp and string data columns is set to On or Auto.
    column name column type parse strings
    time Timestamp Auto
    sym Symbol Auto
    price Float Auto
    volume Integer Auto

    I want to learn more about schemas.

  4. Define the table name to upsert the data to. Select the database and table to write the data to from the dropdown menu; this will be the same database and table from the schema creation step.

    Writer Node

    • Database: databasename
    • Table: pricedata
    • Write Direct to HDB: unchecked
    • Deduplicate stream: checked

    I want to learn more about databases

  5. Open the Pipeline

  6. Save the Pipeline

  7. Deploy the database and associated ACS pipeline.

  8. Query the data. Open the SQL tab. Define the Output variable as "pricedata".

    SQL query

    SELECT * FROM pricedata
    
  9. Get Data

I want to learn more about Microsoft Azure Authentication