The Netflix TechBlog

NOVEMBER 15, 2022

Rendering is the final step in the VFX creation process, and processing on a render farm often can take several hours to complete just a single frame of a show, even when this process runs on the latest high-end hardware. This program is just one example of the many ways Netflix strives to entertain the world.

Cloud 282

Cloud Entertainment AWS Infrastructure 282

Dynatrace

AUGUST 11, 2022

Cloud providers then manage physical hardware, virtual machines, and web server software management. FaaS vs. monolithic architectures. Monolithic architectures were commonplace with legacy, on-premises software solutions. Consider a monolithic application, for example, designed to perform a host of functions.

Development 256

Development Serverless Best Practices Lambda 256

Percona

JUNE 22, 2023

Defining high availability In general terms, high availability refers to the continuous operation of a system with little to no interruption to end users in the event of hardware or software failures, power outages, or other disruptions. If a primary server fails, a backup server can take over and continue to serve requests.

Availability 105

Availability Database Open Source Hardware 105

ACM Sigarch

AUGUST 3, 2023

(Editor’s Note: This post was submitted as a rebuttal to Andrew Chien’s July 24 SIGARCH Blog Post ) The recent post “ Why Embodied Carbon is a poor Architecture Design metric, and Operational Carbon remains an important Problem ” by Prof. Unlike Prof. Chien’s Arguments Against Embodied Carbon Prof. Therefore, the $1.4B

Energy 98

Energy Architecture Hardware Metrics 98

KeyCDN

FEBRUARY 22, 2023

PWAs are designed to work offline, be fast, and provide a seamless user experience across different devices. Motion UI Motion UI is a design trend involving animation and other interactive elements to create a more dynamic and engaging user experience. They thus adapt to the user's browser, screen size, and device specifications.

Artificial Intelligence 111

Artificial Intelligence Development Serverless Website 111

ACM Sigarch

APRIL 8, 2019

Computer architecture is an important and exciting field of computer science, which enables many other fields (eg. For those of us who pursued computer architecture as a career, this is well understood. In most curriculums, undergrad students do not have much exposure to computer architecture. Why is that? Lack of Exposure.

Architecture 74

Architecture Open Source Hardware Software Engineering 74

Percona

APRIL 17, 2023

The swap issue is explained in the excellent article by Jeremy Cole at the Swap Insanity and NUMA Architecture. For example, on a 128GB memory host, this can allow up to 38.4GB of dirty pages. There is an issue with this, which causes the OS to swap even with memory available. The background ratio won’t kick in until 12.8GB.

Best Practices 52

Best Practices Design Tuning Database 52

Adrian Cockcroft

JANUARY 17, 2022

photo taken by Adrian Cockcroft A year ago I did a talk at re:Invent called Architecture Trends and Topics for 2021 , so I thought it was worth seeing how they played out and updating them for the coming year. There were five trends and topics for 2021, Serverless First, Chaos Engineering, Wardley Mapping, Huge Hardware, Sustainability.

Serverless 98

Serverless Hardware AWS Architecture 98

ACM Sigarch

FEBRUARY 23, 2018

In his book “ Misbehaving: The Making of Behavioral Economics ”, he cites example after example of how human behavior does not follow economic expectations based on rational behavior, resulting in flawed economic models . I had a professor in grad school who used to joke that all architecture is reinvented every 5 years.

Architecture 47

Architecture Cache Hardware Virtualization 47

The Morning Paper

JUNE 25, 2019

Designing far memory data structures: think outside the box Aguilera et al., Therefore, if we want to make full use of one-sided far memory, we need to think carefully about the design of our data structures to make that access efficient. This makes it challenging to design effective far memory data structures. A worked example.

Design 80

Design Cache Hardware Scalability 80

ACM Sigarch

JULY 17, 2023

We then discuss defects observed, the role of test and design to address this problem, and identify research opportunities. An example of a specification is the correct operation of the hardware of a microprocessor. An example protection technique that can provide this capability is an Error Correcting Code (ECC).

Energy 52

Energy Hardware Best Practices Architecture 52

ACM Sigarch

APRIL 26, 2018

APU: Accelerated Processing Unit is the AMD’s Fusion architecture that integrates both CPU and GPU on the same die. APU: Accelerated Processing Unit is the AMD’s Fusion architecture that integrates both CPU and GPU on the same die. BPU: Brain Processing Unit is the design of the AI chips by Horizon Robotics.

Hardware 99

Hardware Artificial Intelligence Architecture Google 99

ACM Sigarch

OCTOBER 25, 2019

When it comes to hardware support to mitigate software security issues, there is a significant gap between what is available in products today and known solutions. A History of Architecture Support for Security. The figure above provides a timeline of architectural support for practical defenses, as found in commercial products.

Hardware 98

Hardware Virtualization Architecture Games 98

Dynatrace

APRIL 5, 2021

This is where Lambda comes in: Developers can deploy programs with no concern for the underlying hardware, connecting to services in the broader ecosystem, creating APIs, preparing data, or sending push notifications directly in the cloud, to list just a few examples. How does AWS Lambda work? Many events can trigger a lambda function.

Lambda 184

Lambda AWS Serverless Hardware 184

Percona

JULY 7, 2023

Unfortunately, using certain open source database software as part of an HA architecture can present significant challenges. Where a high availability design once worked well, it can no longer keep up with more complex requirements. Despite all its upside, PostgreSQL software presents such challenges. there cannot be high availability.

Availability 82

Availability Open Source Architecture Database 82

The Morning Paper

NOVEMBER 15, 2020

With more nodes and more coordination comes more complexity, both in design and operation. Today’s paper choice is a wonderful example of pushing the state of the art on a single server. Today’s paper choice is a wonderful example of pushing the state of the art on a single server. IDS/IPS requirements. Back of the envelope.

Servers 128

Servers Hardware Latency Design 128

Testlodge

APRIL 14, 2021

A software tester’s work life includes making sure that systems that have been designed to achieve specific results behave exactly as they should. These systems are a combination of different hardware and software which have been configured to perform the desired task. An Example. Types of Configuration Testing.

Testing 89

Testing Hardware Virtualization Operating System 89

Scalegrid

JANUARY 8, 2024

Discover key insights and strategic advice in our article, designed to steer you toward the best cloud solution that fits your company’s priorities. For example, an application could use on-premises databases while also having access to ‘cloud bursting’ capabilities during times of high demand. But what do these entail?

Strategy 147

Strategy Cloud Scalability Artificial Intelligence 147

The Morning Paper

MAY 12, 2019

An open-source benchmark suite for microservices and their hardware-software implications for cloud & edge systems Gan et al., A typical architecture diagram for one of these services looks like this: Suitably armed with a set of benchmark microservices applications, the investigation can begin! Hardware implications.

Open Source 61

Open Source Hardware Benchmarking Systems 61

Dynatrace

OCTOBER 28, 2021

But with the complexity that comes with digital transformation and cloud-native architecture, teams need a way to make sure applications can withstand the “chaos” of production. To limit the cost of uncovering application vulnerabilities, organizations should to avoid tests that overrun the designated blast radius.

Engineering 231

Engineering Entertainment Cloud Infrastructure 231

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. Programmers continue to write applications in them, and they continue to evolve: the just approved C++20 standard is the latest example.

C++ 108

C++ Benchmarking Hardware Architecture 108

High Scalability

NOVEMBER 14, 2018

Five years ago when Google published The Datacenter as a Computer: Designing Warehouse-Scale Machines it was a manifesto declaring the world of computing had changed forever. The world is still changing, so Google published a new edition: The Datacenter as a Computer: Designing Warehouse-Scale Machines, Third Edition.

Design 54

Design Education Google Internet 54

High Scalability

OCTOBER 24, 2022

Each cloud-native evolution is about using the hardware more efficiently. Examples of traditional natural monopolies are dams, railroads, power plants, water supply, and sewer services. You can even go old school and use non cloud-native architectures. For example, AWS created Nitro. So, the cloud is not a monopoly.

Cloud 187

Cloud Hardware Infrastructure Innovation 187

Dynatrace

SEPTEMBER 30, 2021

It requires purchasing, powering, and configuring physical hardware, training and retaining the staff capable of servicing and securing the machines, operating a data center, and so on. They need enough hardware to serve their anticipated volume and keep things running smoothly without buying too much or too little. Reduced cost.

Cloud 169

Cloud Traffic Best Practices Strategy 169

The Netflix TechBlog

AUGUST 30, 2021

Complementing the hardware is the software on the RAE and in the cloud, and bridging the software on both ends is a bi-directional control plane. For example, when running tests, the state of the device will change from “available for testing” to “in test.” In this blog post, we will focus on the latter feature set.

Latency 213

Latency Traffic Transportation Hardware 213

The Netflix TechBlog

FEBRUARY 9, 2021

Before designing a solution it’s important to understand the main product requirements for such a feature: The content needs to be new, relevant, and regional (not all countries have the same catalogue). To support the demands of our stakeholders we designed a Domain Specific Language (DSL) that we call an asset generation recipe.

Engineering 165

Engineering Storage Latency Entertainment 165

The Netflix TechBlog

JUNE 4, 2019

Because microprocessors are so fast, computer architecture design has evolved towards adding various levels of caching between compute units and the main memory, in order to hide the latency of bringing the bits to the brains. As an illustrative example, let’s consider a toy instance of 16 hyperthreads.

Cache 251

Cache Latency Airlines Logistics 251

The Netflix TechBlog

FEBRUARY 25, 2021

For example, detecting frame boundaries in an AV1 video stream requires being able to parse so-called Open Bitstream Units (OBU) and identifying Temporal Delimiters OBU. Simplified architecture of a streaming preparation pipeline A key feature that our members rightfully deserve when playing audio, video, and timed text is synchronization.

Technology 181

Technology Technology Open Source Media 181

The Netflix TechBlog

DECEMBER 23, 2020

Titus internally employs a cellular bulkhead architecture for scalability, so the fleet is composed of multiple cells. Many bulkhead architectures partition their cells on tenants, where a tenant is defined as a team and their collection of applications. Let’s go back to that kernel code example earlier. User Namespaces.

Media 286

Media Metrics Systems Processing 286

All Things Distributed

MARCH 11, 2016

The expectation was that with each order or two of magnitude, we would need to revisit and revise the architecture to make sure we could address the issues of scale. We needed to build such an architecture that we could introduce new software components without taking the service down. Primitives not frameworks. your resource usage.

AWS 144

AWS Hardware Retail Virtualization 144

Smashing Magazine

OCTOBER 11, 2022

All industries have a part to play, and that includes web design and development. For example, changing our database architecture to be more efficient could save on server costs. On design systems, UX, web performance and CSS/JS. Unfortunately, with some exceptions, that’s generally not the case. Accreditation.

Strategy 124

Strategy Development Energy Website 124

The Morning Paper

MAY 19, 2019

Consider for example a Redis server maintaining a sorted array in memory… … an RPC to add a new entry may incur multiple TLB misses, stalling the core for a few µs while new translations are installed. At one extreme we have a ‘16 x 1’ architecture with 16 queues each with one associated processing unit.

Latency 80

Latency Hardware Network Architecture 80

Percona

SEPTEMBER 1, 2023

To address these challenges, architects must design robust and scalable MongoDB databases and adopt appropriate sharding strategies that can efficiently handle increasing workloads while ensuring continuous availability. Learn more: View our webinar on How to Scale with MongoDB. example.net:27019>" }, { _id : 1, host : "<cfg2.example.net:27019>"

Best Practices 70

Best Practices Traffic Database Servers 70

Sutter's Mill

NOVEMBER 11, 2023

Here is an example from the paper that will now be legal: template <typename. Instead, you want a library that is tuned for your target hardware architecture and ready for par_unseq vectorized algorithms, for blazing speed. T> constexpr auto first_plus_last(T. values) -> T.[0] [0] { return T.[0](values.[0]

C++ 107

C++ Lambda Network Programming 107

The Morning Paper

NOVEMBER 10, 2019

It’s been clear for a while that software designed explicitly for the data center environment will increasingly want/need to make different design trade-offs to e.g. general-purpose systems software that you might install on your own machines. But wow, I didn’t think we’d be at the point yet where we’d be abandoning TCP/IP!

Network 92

Network Transportation Latency Entertainment 92

Scalegrid

JANUARY 5, 2024

It was initially designed to mitigate the limitations of file management systems, including slow operations, inadequate security, and substantial data redundancy. It comprises a collection of interrelated data and a set of software tools that aid in the access, processing, and management of data.

Efficiency 147

Efficiency Storage Database Scalability 147

ACM Sigarch

OCTOBER 4, 2018

This is the first post in a series of posts on different approaches to systems security especially as they apply to hardware and architectural security. For example, consider a database system that is distributed across a data center, or even across data centers. These two, yet again, are each useful in different circumstances.

Systems 47

Systems Hardware Storage Healthcare 47