Facilitating the Spread of Knowledge and Innovation in Professional Software Development

Write for InfoQ


Choose your language

InfoQ Homepage News Next Generation of Data Movement and Processing Platform at Netflix

Next Generation of Data Movement and Processing Platform at Netflix

Netflix engineering recently published in a tech blog how they used data mesh architecture and principles as the next generation of data platform and processing to unleash more business use cases and opportunities.

Data mesh is the new paradigm shift in data management that enables users to easily import and use data without transporting it to a centralized location like a data lake. It focuses on decentralization and distributing data ownership in different business domains. Each domain manages and governs its data as a product, making it discoverable and trusted based on required business service level objectives (SLOs).

The data mesh architecture in Netflix is composed of two main layers. One is the control plane (controllers) and the other is the data plane (data pipelines). It is defined in the blog post simply as: 

The controller receives user requests and deploys and orchestrates pipelines. Once deployed, the pipeline performs the actual heavy lifting data processing work. Provisioning a pipeline involves different resources. The controller delegates the responsibility to the corresponding microservices to manage their life cycle.

The following diagram shows the high-level architecture of the data mesh in Netflix : 

Netflix high-level data mesh architecture

The main operations are done in pipelines. Pipeline reads data from various resources, applies algorithms, and transports them to the destination. The controllers are responsible for figuring out the resources associated with the pipelines and calculating the correct configuration. The following diagram shows an example of the pipeline architecture:


Pipelines high-level architecture

Data sources are domain data related to each business unit and processed by different processors in the system. Engineers are mainly using Apache Flink for real-time data processing. Connectors are the main element to initiate data transfer. They monitor data sources and produce change data capture (CDC) events to the data mesh. Apache Kafka is acting as the main component for data transporters in data mesh. Data schema and catalog are very important in providing searchability and visibility of the data among different business domains. Netflix uses Apache Avro as the standard schema among domains.

Many businesses have already started to shift into this paradigm based on their needs. Chris Damour, one of the readers mentioned in the blog post comments :

We have also implemented the features to track the data lineage so that our users can have a better picture of the overall data usage. I'd love to see a blog on that, data provenance has been a real struggle in our data meshes, we get the "same".

Enterprises are now planning and shifting their data platform based on this architecture including Intuit and Zalando. Many cloud providers like Amazon, Google, and Microsoft are also providing data mesh solutions and services for their customers.

About the Author

Rate this Article