Build streaming data pipelines in minutes, respond to events in seconds.

Furnace allows you to spend less time designing and managing infrastructure, leaving you more time to focus on doing great things with your data.

A platform that Dev, Sec and Ops love.

An opinionated yet flexible streaming data pipeline solution.

Furnace makes it easy to build highly effective teams that can respond to a continuously changing landscape, using best practices and a modern technology stack.

Get Started Now
Platform Agnostic
Platform Agnostic

A single deployment experience across multiple platforms

Platform Agnostic
Deploys Native

Furnace utilises native infrastructure and functionality to maximise efficiency

Platform Agnostic
Minimal Cost

By scaling services as required Furnace ensures you're only using what you need

Platform Agnostic
Everything As Code

Describe your entire data pipeline in code and have Furnace handle the deployment into multiple environments

Platform Agnostic
GitOps

Furnace aligns itself with the emerging GitOps methodology, where Git is the single source of truth bringing many benefits to modern development teams.

Platform Agnostic
We're Serverless

Providing huge efficiency wins, no longer do you need to stand up stacks of infrastructure that costs even whilst idle.

Furnace Constructs

Furnace provides a set of simple contructs that are the building blocks of your data pipeline.

Furnace Illustration

Sources

Defines source of data, usually this is stream from Apache Kafka or AWS Kinesis but can be anything you define. We provide some standard sources out of the box.

Taps

Taps connect to a source. Their job is to parse and normalise data into a common format. A default set of Taps are provided and new Taps can be created by simply writing a serverless function.

Pipelines

Pipelines create a linear path for data to flow through a chosen set of functions. Pipelines are connected to Taps and a Tap can feed multiple Pipelines.

Sinks

A Sink is where your data arrives after it exits a Pipeline. A sink could be a data lake or storage bucket. Multiple Pipelines can feed into a Sink and a Pipeline can feed into multiple Sinks.

Resources

Resources are used to initiate resources native to the environment in which Furnace is being deployed.

Actions

Once data has been processed by your pipelines, Actions make use of the structured data react and automate tasks in real-time.

Stacks

A Stack is comprised of one or more end to end data flows into a logical container. A Stack can have multiple environments (Dev, Staging, Production).

See Furnace in Action. Zero to Hero in just 5 minutes.

play

Reference Architecture

Furnace targets a number of uses cases across Security, IoT and many other streaming data needs. Over the coming months we'll be releasing more exmples of how to use Furnace, starting with a popular use on Amazon Web Services.


Sources

Our AWS reference application makes use of VPC Flow Logs via CloudWatch Logs to anaylse traffic flowing over a Virtual Provide Cloud.

Tap

The Tap decodes the data, normalises it into a common format, and pushes it into AWS Kinesis for further processing by the Pipeline

Pipeline

As the Pipeline receives events from the Tap, they are sent through a series of steps, adding further AWS metadata and Geo Location of IP addresses.

Sink

AWS provides various managed Big Data storage and analysis services including ElasticSearch, RedShift and even S3. Furnace integrates into many of the AWS services.

Join our Community

We'd love you to help us build Furnace, in the form of Pull Requests, feedback from your journey or features you'd like to see.

The Furnace Blog

Back in 2016, our team began researching the technologies in which Security Operations Center’s were using to protect the broad range of customers they monitor. Security Information and Event Management (SIEM)...

Follow this guide for step by step guidance on building a Furnace stack on AWS from a CloudWatch Logs Stream thought a processing Pipeline and into an ElasticSearch Sink.

Our Integrations

Furnace aims to add native support for all major Cloud Providers, we'll be adding more in the coming months.

AWSGitHubPulumi

Furnace is Open Source and available now

Milestones

We mean business and are following an aggressive roadmap into 2019

1

Initial Release

We're launching with support for the AWS Cloud and GitHub as a Source Repository with tight integrations into both.

2

Additional Languages and Cloud Providers

We're going to add support for writing modules in Python and Go as well as the ability to run Furnace on Microsoft Azure and Google Cloud Platform.

3

SIEM Enablement Use Case

Constructs to connect cloud based applications to legacy on premise platforms.