article thumbnail

Scaling Media Machine Learning at Netflix

The Netflix TechBlog

Our goal in building a media-focused ML infrastructure is to reduce the time from ideation to productization for our media ML practitioners. We accomplish this by paving the path to: Accessing and processing media data (e.g. We accomplish this by paving the path to: Accessing and processing media data (e.g.

Media 290
article thumbnail

For your eyes only: improving Netflix video quality with neural networks

The Netflix TechBlog

Recently, we added another powerful tool to our arsenal: neural networks for video downscaling. In this tech blog, we describe how we improved Netflix video quality with neural networks, the challenges we faced and what lies ahead. How can neural networks fit into Netflix video encoding?

Network 292
Insiders

Sign Up for our Newsletter

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

Trending Sources

article thumbnail

How To Implement Video Information and Editing APIs in Java

DZone

It must be said that this video traffic phenomenon primarily owes itself to modernizations in the scalability of streaming infrastructure, which simply weren’t present fifteen years ago.

article thumbnail

Building In-Video Search

The Netflix TechBlog

Engineering and Infrastructure Our trained model gives us a text encoder and a video encoder. Video embeddings are precomputed on the shot level, stored in our media feature store , and replicated to an elastic search cluster for real-time nearest neighbor queries. However, shot segmentation from a full-length movie is CPU-intensive.

Media 225
article thumbnail

Rebuilding Netflix Video Processing Pipeline with Microservices

The Netflix TechBlog

When Reloaded was designed, we focused on a single use case: converting high-quality media files (also known as mezzanines) received from studios into compressed assets for Netflix streaming. To achieve this, Cosmos was developed as a computing platform for workflow-driven, media-centric microservices.

article thumbnail

Designing Instagram

High Scalability

User Feed Service, Media Counter Service) read the actions from the streaming data store and performs their specific tasks. media search index, locations search index, and so forth) in future. For instance, the LikeEvent can be read by the Media Counter Service and is used to update the media count in the data storage.

Design 334
article thumbnail

OpenShift vs. Kubernetes: Understanding the differences

Dynatrace

Container orchestration allows an organization to digitally transform at a rapid clip without getting bogged down by slow, siloed development, difficult scaling, and high costs associated with optimizing application infrastructure. Networking. Kubernetes provides a basic networking model. Updates and support.