article thumbnail

Migrating Critical Traffic At Scale with No Downtime?—?Part 1

The Netflix TechBlog

Migrating Critical Traffic At Scale with No Downtime — Part 1 Shyam Gala , Javier Fernandez-Ivern , Anup Rokkam Pratap , Devang Shah Hundreds of millions of customers tune into Netflix every day, expecting an uninterrupted and immersive streaming experience. This approach has a handful of benefits.

Traffic 339
article thumbnail

WebP Caching has Landed!

KeyCDN

We’re happy to announce that WebP Caching has landed! How Does WebP Caching Work? jpg in the example above) will not change. Below is an example for Nginx that manages the delivery of WebP assets on your origin server. Enable the Feature for your Zones Cache Key WebP can be enabled for all Pull Zones.

Cache 81
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

Supporting Diverse ML Systems at Netflix

The Netflix TechBlog

Example use case: Content Knowledge Graph Our knowledge graph of the entertainment world encodes relationships between titles, actors and other attributes of a film or series, supporting all aspects of business at Netflix. For many of our applications, model explainability matters. ETL workflows), as well as downstream (e.g.

Systems 226
article thumbnail

Crucial Redis Monitoring Metrics You Must Watch

Scalegrid

Effective management of memory stores with policies like LRU/LFU proactive monitoring of the replication process and advanced metrics such as cache hit ratio and persistence indicators are crucial for ensuring data integrity and optimizing Redis’s performance. Cache Hit Ratio The cache hit ratio represents the efficiency of cache usage.

Metrics 130
article thumbnail

The road to observability with OpenTelemetry demo part 1: Identifying metrics and traces

Dynatrace

But its underlying goal is quite humble and straightforward: it wants to enable you to observe an IT system (for example, a web application, infrastructure, or services) and gain insight to its behavior, such as performance, error rates, hot spots of executed instructions in code, and more. Those are prime candidates for their own spans.

Metrics 182
article thumbnail

Seamlessly Swapping the API backend of the Netflix Android app

The Netflix TechBlog

This allows the app to query a list of “paths” in each HTTP request, and get specially formatted JSON (jsonGraph) that we use to cache the data and hydrate the UI. For example, the artwork service is separate from the video metadata service, but we need the data from both in the detail key. Replay Testing Enter replay testing.

Latency 233
article thumbnail

Compress objects, not cache lines: an object-based compressed memory hierarchy

The Morning Paper

Compress objects, not cache lines: an object-based compressed memory hierarchy Tsai & Sanchez, ASPLOS’19. So while you probably won’t be using Zippads in practice anytime soon, it’s a wonderful example of what’s possible when you’re prepared to take a fresh look at “the way we’ve always done things.”

Cache 61