The Netflix TechBlog

SEPTEMBER 18, 2024

In this post, we dive deep into how Netflix’s KV abstraction works, the architectural principles guiding its design, the challenges we faced in scaling diverse use cases, and the technical innovations that have allowed us to achieve the performance and reliability required by Netflix’s global operations.

Latency 256

Latency Storage Cache Efficiency 256

The Netflix TechBlog

OCTOBER 8, 2024

In previous blog posts, we introduced the Key-Value Data Abstraction Layer and the Data Gateway Platform , both of which are integral to Netflix’s data architecture. Handling Bursty Traffic : Managing significant traffic spikes during high-demand events, such as new content launches or regional failovers.

Latency 235

Latency Storage Traffic Tuning 235

The Netflix TechBlog

MARCH 10, 2023

These applications are built on a microservices architecture, and the Asset Management Platform provides asset management to those dozens of services for various asset types. To avoid the ES query for the list of indices for every indexing request, we keep the list of indices in a distributed cache.

Strategy 261

Strategy Cache Storage Analytics 261

The Netflix TechBlog

OCTOBER 16, 2019

By Ammar Khaku Introduction In a microservice architecture such as Netflix’s, propagating datasets from a single source to multiple downstream destinations can be challenging. This post is a high level overview of the design and architecture of Gutenberg. A publisher publishes to a topic and consumers consume from a topic.

Cache 189

Cache Architecture Metrics Java 189

Scalegrid

MARCH 28, 2024

Key Takeaways Redis offers complex data structures and additional features for versatile data handling, while Memcached excels in simplicity with a fast, multi-threaded architecture for basic caching needs. Introduction Caching serves a dual purpose in web development – speeding up client requests and reducing server load.

Cache 130

Cache Storage Architecture Scalability 130

Dynatrace

JANUARY 23, 2023

Here’s the update: Improve architectural design to eliminate SSO bottleneck risk [In progress] Security and access are critical aspects of our architecture, and as such, there are many areas we’re looking to improve. Hopefully never.) This has been completed.

Social Media 130

Social Media Cache Traffic Media 130

Smashing Magazine

JUNE 29, 2021

Without build optimizations (incremental builds, caching, we will get to those soon) this will eventually become unmanageable as well — think about going through all images in a website: resizing, deleting, and/or creating new files over and over again. Jamstack general service architecture ( Large preview ).

Cache 139

Cache Website Strategy Traffic 139

The Netflix TechBlog

DECEMBER 17, 2019

Nonetheless, we found a number of limitations that could not satisfy our requirements e.g. stalling the processing of log events until a dump is complete, missing ability to trigger dumps on demand, or implementations that block write traffic by using table locks. Blocking write traffic by locking tables. DBLog High Level Architecture.

Database 216

Database Traffic Transportation Open Source 216

Percona

MARCH 20, 2023

When designing an architecture, many components need to be considered before deciding on the best solution. In this context, features like filtering, firewalling, or caching are redundant and may consume resources that could be allocated to scaling.

Games 141

Games Latency Traffic Cache 141

The Netflix TechBlog

DECEMBER 17, 2019

Nonetheless, we found a number of limitations that could not satisfy our requirements e.g. stalling the processing of log events until a dump is complete, missing ability to trigger dumps on demand, or implementations that block write traffic by using table locks. Blocking write traffic by locking tables. DBLog High Level Architecture.

Database 202

Database Traffic Transportation Open Source 202

High Scalability

JULY 20, 2018

That’s mapping applications to the specific architectural choices. The third wing of the architecture piece is the “domain specific system-on-chip.” That means multiple data indirections mean multiple cache misses. They are very expensive. This is where your performance goes.

Internet 121

Internet Internet Scalability Automotive 121

The Netflix TechBlog

MARCH 5, 2019

the order of the rows on your Netflix home page, issuing content licenses when you click play, finding the Open Connect cache closest to you with the content you requested, and many more). Can we adjust our auto-scaling policies to be more efficiency without risking our availability during traffic spikes?

Infrastructure 184

Infrastructure Cloud Scalability AWS 184

Smashing Magazine

DECEMBER 23, 2020

The service workers enable the offline usage of the PWA by fetching cached data or informing the user about the absence of an Internet connection. Application shell architecture. When developing a PWA, you can cache the application shell’s resources and assets in the browser. Cached content with IndexedDB.

Cache 136

Cache Internet Internet Google 136

Dynatrace

SEPTEMBER 17, 2019

REDIS for caching. MaaSS for Cloud Architects: Deployment and Architecture Validations. Thanks to PurePath, architects can validate how transactions flow from service-to-service and how traffic gets routed through service mashes (AWS App Mesh, Istio, Linkerd) or proxies. NGINX as an API Gateway. 1 Validate Deployment.

AWS 170

AWS Monitoring Ecommerce Lambda 170

IO River

NOVEMBER 2, 2023

CDNs cache content on edge servers distributed globally, reducing the distance between users and the content they want.‍CDNs use load-balancing techniques to distribute incoming traffic across multiple servers called Points of Presence (PoPs) which distribute content closer to end-users and improve overall performance.

Architecture 52

Architecture Cache Performance Latency 52

IO River

NOVEMBER 2, 2023

CDNs cache content on edge servers distributed globally, reducing the distance between users and the content they want.‍CDNs CDNs use load-balancing techniques to distribute incoming traffic across multiple servers called Points of Presence (PoPs) which distribute content closer to end-users and improve overall performance.Â

Architecture 52

Architecture Cache Performance Latency 52

Percona

JUNE 22, 2023

Load balancing : Traffic is distributed across multiple servers to prevent any one component from becoming overloaded. Load balancers can detect when a component is not responding and put traffic redirection in motion. When planning your database HA architecture, the size of your company is a great place to start to assess your needs.

Availability 132

Availability Database Open Source Hardware 132

O'Reilly Software

DECEMBER 11, 2017

A message-based microservices architecture offers many advantages, making solutions easier to scale and expand with new services. The asynchronous nature of interservice interactions inherent to this architecture, however, poses challenges for user-initiated actions such as create-read-update-delete (CRUD) requests on an object.

Architecture 68

Architecture Government Latency Efficiency 68

Software Architecture

AUGUST 24, 2017

Memcached is very good at handling high traffic websites. Redis can not handle heavy traffic on read/write. Redis vs Memcached was originally published in Software Architecture on Medium, where people are continuing the conversation by highlighting and responding to this story.

Traffic 100

Traffic Software Architecture Website Efficiency 100

InfoQ

JULY 3, 2023

LinkedIn introduced Couchbase as a centralized caching tier for scaling member profile reads to handle increasing traffic that has outgrown their existing database cluster. The new solution achieved over 99% hit rate, helped reduce tail latencies by more than 60% and costs by 10% annually. By Rafal Gancarz

Cache 85

Cache Latency Traffic Database 85

ACM Sigarch

OCTOBER 12, 2020

The focus of most published research in architecture is on applications implemented in high-performance, “ close-to-the-metal” languages essentially developed before computers got fast. Are caches large enough for this code? Despite their age, these languages are far from dead! What about branch prediction tables?

C++ 108

C++ Benchmarking Hardware Architecture 108

The Morning Paper

OCTOBER 4, 2020

Our analysis suggests that the wireline paths, upper-layer protocols, computing and radio hardward architecture need to co-evolve with 5G to form an ecosystem, in order to fully unleash its potential. This is a feature of the NSA architecture which requires dropping off of 5G onto 4G, doing a handover on 4G, and then upgrading to 5G again.

Energy 130

Energy Latency Performance Network 130

All Things Distributed

JANUARY 18, 2012

s web-based applications often encounter database scaling challenges when faced with growth in users, traffic, and data. Behind the scenes, Amazon DynamoDB automatically spreads the data and traffic for a table over a sufficient number of servers to meet the request capacity specified by the customer.

Scalability 162

Scalability Database Ecommerce Latency 162

Dotcom-Montior

JUNE 25, 2020

This project implements multiple techniques and tactics to censor China’s internet and controls the internet gateways to analyze, filter, and manipulate the internet traffic between inside and outside of China. It’s a complex architecture with multiple layers of controlled tricks to block access to websites. Blocking IP Access.

Website 97

Website Internet Internet Performance 97

Jos

JUNE 2, 2018

UI/UX : There is usually a designer and/or UX person that sets the look & feel and information architecture. When they dug into the data, they found that the reason load times had increased was that they got a lot more traffic from Africa after doing the optimizations. Testing : Well, we should write some tests, right?

Performance 237

Performance Speed Google Media 237

The Netflix TechBlog

OCTOBER 18, 2019

It featured three relevant talks from LinkedIn, Netflix and Facebook, and a platform architecture overview talk from first time participant Dropbox. In particular, he talked about the misattribution potential in a complex microservice architecture where often intermediary results are cached.

Infrastructure 129

Infrastructure Metrics Architecture Efficiency 129

Smashing Magazine

AUGUST 3, 2023

It increases our visibility and enables us to draw a steady stream of organic (or “free”) traffic to our site. While paid marketing strategies like Google Ads play a part in our approach as well, enhancing our organic traffic remains a major priority. The higher our organic traffic, the more profitable we become as a company.

Performance 78

Performance Cache Traffic Metrics 78

Dotcom-Montior

MARCH 27, 2022

This project implements multiple techniques and tactics to censor China’s internet and controls the internet gateways to analyze, filter, and manipulate the internet traffic between inside and outside of China. It’s a complex architecture with multiple layers of controlled tricks to block access to websites. Blocking IP Access.

Website 82

Website Internet Internet Performance 82

Scalegrid

FEBRUARY 8, 2024

Distributed Storage Architecture Distributed storage systems are designed with a core framework that includes the main system controller, a data repository for the system, and a database. This strategy reduces the volume needed during retrieval operations.

Storage 130

Storage Systems Big Data Azure 130

Smashing Magazine

JUNE 25, 2020

Today, we’ll address storing and serving files for both single-server and scalable deployments while considering factors like compression, caching, and availability. This strategy is very simple and closely resembles the development environment, but cannot handle large or inconsistent amounts of traffic effectively. In Conclusion.

Media 77

Media Cache Storage AWS 77

The Morning Paper

JUNE 25, 2019

A far memory data structure has: far data in far memory, containing the core content of the data structure data caches at clients algorithms for operations. Processor caches can help to hide local accesses too, but not remote accesses. Clients cache the entire tree, but not the hash tables. Refreshable vectors. A worked example.

Design 80

Design Cache Hardware Scalability 80

Alex Russell

MARCH 6, 2021

A then-representative $200USD device had 4-8 slow (in-order, low-cache) cores, ~2GiB of RAM, and relatively slow MLC NAND flash storage. The fastest Androids predictably remain 18-24 months behind, owing to cheapskate choices about cache sizing by Qualcomm, Samsung Semi, and all the rest. The Moto G4 , for example.

Performance 127

Performance Network Mobile Metrics 127

The Symphonia

MAY 22, 2018

I started writing “ Serverless Architectures ” in May 2016. I was a little restricted in my thinking the first time around and I’ve come to see FaaS as something not quite stateless, since caching state in a Lambda instance that might stick around for 5 hours is a perfectly reasonable idea. I thought a few folks might be interested.

Serverless 40

Serverless Architecture Lambda Azure 40

Scalegrid

JANUARY 5, 2024

However, some challenges may arise when scaling a DBMS, such as improper traffic distribution, inefficient database management, and performance issues. By implementing data abstraction techniques, these challenges can be addressed more effectively.

Efficiency 147

Efficiency Storage Database Scalability 147

The Netflix TechBlog

OCTOBER 18, 2019

It featured three relevant talks from LinkedIn, Netflix and Facebook, and a platform architecture overview talk from first time participant Dropbox. In particular, he talked about the misattribution potential in a complex microservice architecture where often intermediary results are cached.

Infrastructure 100

Infrastructure Metrics Architecture Efficiency 100

Abhishek Tiwari

NOVEMBER 3, 2018

When it comes to innovation, most of CMS solutions are constrained by their legacy architecture (read strong coupling between content management and content presentation) which makes it difficult to serve content to new types of emerging channels such as apps and devices. Eventually, we decided to move them to Jekyll.

Systems 63

Systems Cache Website Network 63

IO River

NOVEMBER 2, 2023

They cache static content and enable lightning-fast delivery around the globe.This symbiosis reduces server load, boosts loading times, and ensures efficient content distribution. Content Delivery Networks (CDNs), web browsers, and proxy servers can store static files in their caches. For example, consider tools like ChatGPT.

Cache 52

Cache Social Media Website Performance Website 52