Remove Analytics Remove Database Remove Latency Remove Video
article thumbnail

Rebuilding Netflix Video Processing Pipeline with Microservices

The Netflix TechBlog

The Netflix video processing pipeline went live with the launch of our streaming service in 2007. This architecture shift greatly reduced the processing latency and increased system resiliency. For example, in Reloaded the video quality calculation was implemented inside the video encoder module.

article thumbnail

Scalable Annotation Service?—?Marken

The Netflix TechBlog

An example for storing both time and space based data would be an ML algorithm that can identify characters in a frame and wants to store the following for a video In a particular frame (time) In some area in image (space) A character name (annotation data) Pic 1 : Editors requesting changes by drawing shapes like the blue circle shown above.

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

Designing Instagram

High Scalability

Generating machine learning based personalized recommendations to discover new people, photos, videos, and stories relevant one’s interest. We will use a graph database such as Neo4j to store the information. Additionally, we can use columnar databases like Cassandra to store information like user feeds, activities, and counters.

Design 334
article thumbnail

Procella: unifying serving and analytical data at YouTube

The Morning Paper

Procella: unifying serving and analytical data at YouTube Chattopadhyay et al., That’s hard for many reasons, including the differing trade-offs between throughput and latency that need to be made across the use cases. Oh, and in additional to low latency, “ we require access to fresh data.” VLDB’19.

article thumbnail

Bulldozer: Batch Data Moving from Data Warehouse to Online Key-Value Stores

The Netflix TechBlog

Data scientists and engineers collect this data from our subscribers and videos, and implement data analytics models to discover customer behaviour with the goal of maximizing user joy. The data warehouse is not designed to serve point requests from microservices with low latency.

Latency 243
article thumbnail

Data Reprocessing Pipeline in Asset Management Platform @Netflix

The Netflix TechBlog

Production Use Cases Real-Time APIs (backed by the Cassandra database) for asset metadata access don’t fit analytics use cases by data science or machine learning teams. Production assets operations are performed in parallel with older data reprocessing without any service downtime.

Media 237
article thumbnail

Accelerating Data: Faster and More Scalable ElastiCache for Redis

All Things Distributed

Fast Data is an emerging industry term for information that is arriving at high volume and incredible rates, faster than traditional databases can manage. Three years ago, as part of our AWS Fast Data journey we introduced Amazon ElastiCache for Redis , a fully managed in-memory data store that operates at sub-millisecond latency.