Quickstart

This guide will walk you through the process of setting up Fluvio and then will introduce you to the basic concepts.

Install Fluvio

Fluvio is installed via the Fluvio Version Manager, also known as fvm.

To install fvm, run the following command:

curl -fsS https://hub.infinyon.cloud/install/install.sh | bash

☁️  Downloading fluvio version manager, fvm
   target arch aarch64-apple-darwin
⬇️ Installing fvm
   ...

As part of the initial setup, fvm will also install the Fluvio CLI available in the stable channel as of the moment of installation.

Fluvio is stored in $HOME/.fluvio, with the executable binaries stored in $HOME/.fluvio/bin.

tip

For the best compatibliity on Windows, InfinyOn recommends WSL2

Start a Cluster

Start cluster on you local machine with the following command:

$ fluvio cluster start

tip

To run a cloud cluster, see the Cloud documentation.

Create your first topic

Topics are used to store data and send data streams.

You can create a topic with the following command:

$ fluvio topic create quickstart-topic

Where quickstart-topic is the name of your topic

Produce Records

You can send data (aka produce) to your topic.

Let’s try to produce text to your topic interactively:

$ fluvio produce quickstart-topic
> hello world!
Ok!

Typing anything and then pressing Enter will send a record to your topic.

Press Ctrl+C to exit the interactive producer prompt.

Consume Records

You can read data (aka consume) from your topic.

This command will create a consumer that listens to your topic for new records and then prints it to the screen:

$ fluvio consume quickstart-topic
Consuming records from the end of topic 'quickstart-topic'. This will wait for new records

Note, that hello world! was already sent to the topic, so you will not see it printed to the screen.

To see previously sent data, open a new terminal and add an option to your consume command to request a starting offset with the -B, --beginning <offset> flag.

$ fluvio consume quickstart-topic -B -d
hello world!

Flags:

• -B, --beginning flag is used to specify the starting offset. Defaults to 0
• -d, --disable-continuous closes the consumer connection after all data has been sent.

Start a Connector

InfinyOn offers a growing number of connectors to communicate with external services. In this example, we will be covering the HTTP Source connector. The connector polls data from an HTTP endpoint that returns a random quote every 3 seconds to a topic called quotes.

Create a file and save the following configuration:

# quotes-source-connector.yml
apiVersion: 0.1.0
meta:
  version: 0.3.8
  name: http-quotes
  type: http-source
  topic: quotes
http:
  endpoint: https://demo-data.infinyon.com/api/quote
  interval: 3s

Running the HTTP Connector

We'll use Connector Developer Kit (cdk) to download and run the connector.

$ cdk hub download infinyon/http-source@0.3.8

Run the following command to deploy the connector:

$ cdk deploy start --ipkg infinyon-http-source-0.3.8.ipkg -c quotes-source-connector.yml

Use the following command to see the connector status.

$ cdk deploy list
 NAME         STATUS
 http-quotes  Running

We can monitor new data in the connector’s topic with fluvio consume quotes

$ fluvio consume quotes
Consuming records from 'quotes'
{"quote":"We cannot solve our problems with the same thinking we used when we created them.","by":"Albert Einstein"}
{"quote":"Whatever you are, be a good one.","by":"Abraham Lincoln"}
{"quote":"You can't build a reputation on what you're going to do.","by":"Henry Ford"}
{"quote":"Success is not final, failure is not fatal: It is the courage to continue that counts.","by":"Winston Churchill"}

Transform Records with SmartModules

SmartModules are user-defined functions compiled into WebAssembly and applied to data streaming for inline data manipulation. You can use SmartModules in the producers, consumers, as well as Connectors. InfinyOn has several pre-compiled SmartModules that you can use out of the box. Alternatively, you can build your own.

To list available smartmodules, run:

$ fluvio hub smartmodule list
  SMARTMODULE                              Visibility 
  infinyon/jolt@0.4.1                      public     
  ...

Use jolt SmartModule to turn JSON records into sentences

Let’s download the Smartmodule to our cluster:

$ fluvio hub smartmodule download infinyon/jolt@0.4.1

Check the cluster to ensure it has been successfully downloaded:

$ fluvio smartmodule list
  SMARTMODULE          SIZE     
  infinyon/jolt@0.4.1  589.3 KB 

Next, we’ll create a transform file and test the output.

Create a SmartModule transformation file

SmartModules can be chained together and often require additional parameters. Fluvio uses a YAML file is used to define the transformations.

Create a file named transforms.yaml and paste the following configuration:

#transforms.yml
transforms:
  - uses: infinyon/jolt@0.4.1
    with:
      spec:
        - operation: shift
          spec:
            quote: ""

Jolt is a complex Smartmodule that allows you to perform multiple types of JSON transformations. For additional information, check out the SmartModule Jolt docs.

Test the SmartModule

$  fluvio consume quotes --transforms-file transforms.yaml -T=2
Consuming records from 'quotes' starting 2 from the end of log
"The greatest glory in living lies not in never falling, but in rising every time we fall."
"Simplicity is the ultimate sophistication."

We are consuming the last two quotes topic records and transforming the json into a string. The stream continues consuming as our connectors is still running.

Create a data pipeline with a Connector and a SmartModule

Let’s say we don’t use the authors in the quotes; instead, only the quote represented strings. We can accomplish this result by simply applying the transformation to the connector.

Let’s create a new file add the following configuration:

# string-quotes-source-connector.yml
apiVersion: 0.1.0
meta:
  version: 0.3.8
  name: string-quotes
  type: http-source
  topic: string-quotes
http:
  endpoint: https://demo-data.infinyon.com/api/quote
  interval: 3s
transforms:
  - uses: infinyon/jolt@0.4.1
    with:
      spec:
        - operation: shift
          spec:
            quote: ""

Ready to run the connector:

$ cdk deploy start --ipkg infinyon-http-source-0.3.8.ipkg -c string-quotes-source-connector.yml

Use the following command to see the connector status.

$ cdk deploy list
 NAME           STATUS
 http-quotes    Running
 string-quotes  Running

Let’s take a look at string-quotes

$ fluvio consume string-quotes
Consuming records from 'string-quotes'
"It's not whether you get knocked down, it's whether you get up."
"Honesty is the first chapter in the book of wisdom."

We now have two topics running in parallel and producing different results with a simple SmartModule transformation. When you apply inline transformations, the number of possibilities are virtually endless.

Clean-up Resources

During this tutorial, we’ve created a couple connectors and three topics. Let's clean-up:

cdk deploy shutdown --name http-quotes
cdk deploy shutdown --name string-quotes
fluvio topic delete quotes
fluvio topic delete string-quotes
fluvio topic delete quickstart-topic

Build complex data pipelines with Stateful Dataflows (sdf)

Stateful Dataflows, also known as SDF, allow you to build hierarchical data pipelines with enrichment, window processing, and more.

Checkout the SDF section for more information.