The Morning Paper

SEPTEMBER 3, 2019

SLOG: serializable, low-latency, geo-replicated transactions Ren et al., Strict serializability reduces application code complexity and bugs, since it behaves like a system that is running on a single machine processing transactions sequentially. That’s where SLOG (Serializable LOw-latency, Geo-replicated transactions) comes in.

Latency 70

Latency Processing Benchmarking Systems 70

The Netflix TechBlog

MARCH 6, 2019

—?Mounting object storage in Netflix’s media processing platform By Barak Alon (on behalf of Netflix’s Media Cloud Engineering team) MezzFS (short for “Mezzanine File System”) is a tool we’ve developed at Netflix that mounts cloud objects as local files via FUSE. Encoding is not a one-time process?—?large We have one file?—?the

Media 214

Media Storage Processing Cache 214

CSS Wizardry

OCTOBER 16, 2023

In this post, I’m going to break these processes down into each of: ? Plotted on the same horizontal axis of 1.6s, the waterfalls speak for themselves: 201ms of cumulative latency; 109ms of cumulative download. 4,362ms of cumulative latency; 240ms of cumulative download. Read the complete test methodology. It gets worse.

Cache 291

Cache Latency Strategy Speed 291

DZone

SEPTEMBER 14, 2022

Typically in low-latency development, a trade-off must be made between minimizing latency and avoiding excessive CPU utilization. In a typical application stack, multiple threads are used for servicing events, processing data, pipelining, and so on. Description of the Problem.

Latency 130

Latency Open Source Strategy Design 130

The Netflix TechBlog

MARCH 10, 2023

Hence we built the data pipeline that can be used to extract the existing assets metadata and process it specifically to each new use case. This feature support required a significant update in the data table design (which includes new tables and updating existing table columns). N, first N rows are fetched from the table.

Media 237

Media Traffic Processing Design 237

The Netflix TechBlog

NOVEMBER 3, 2022

We introduce a caching mechanism in the API gateway layer, allowing us to offload processing from singleton leader elected controllers without giving up strict data consistency and guarantees clients observe. cell): Titus Job Coordinator is a leader elected process managing the active state of the system.

Cache 224

Cache Latency Traffic Systems 224

The Netflix TechBlog

SEPTEMBER 24, 2021

By Xiaomei Liu , Rosanna Lee , Cyril Concolato Introduction Behind the scenes of the beloved Netflix streaming service and content, there are many technology innovations in media processing. Packaging has always been an important step in media processing. Uploading and downloading data always come with a penalty, namely latency.

Cloud 237

Cloud Media Storage Cache 237

Dynatrace

JANUARY 31, 2024

The RAG process begins by summarizing and converting user prompts into queries that are sent to a search platform that uses semantic similarities to find relevant data in vector databases, semantic caches, or other online data sources. RAG augments user prompts with relevant data retrieved from outside the LLM.

Cache 199

Cache Azure Infrastructure Monitoring 199

The Netflix TechBlog

MARCH 1, 2021

It supports both high throughput services that consume hundreds of thousands of CPUs at a time, and latency-sensitive workloads where humans are waiting for the results of a computation. The subsystems all communicate with each other asynchronously via Timestone, a high-scale, low-latency priority queuing system.

Serverless 228

Serverless Media Latency Social Media 228

Dynatrace

JUNE 7, 2023

Advanced AI applications using OpenAI services don’t just forward user input to OpenAI models; they also require client-side pre- and post-processing. A typical design pattern is the use of a semantic search over a domain-specific knowledge base, like internal documentation, to provide the required context in the prompt.

Monitoring 159

Monitoring Latency Metrics Azure 159

The Netflix TechBlog

NOVEMBER 3, 2023

Democratizing Stream Processing @ Netflix By Guil Pires , Mark Cho , Mingliang Liu , Sujay Jain Data powers much of what we do at Netflix. On the Data Platform team, we build the infrastructure used across the company to process data at scale. In our last blog post, we introduced “Data Mesh” — A Data Movement and Processing Platform.

Processing 265

Processing Engineering Infrastructure Latency 265

Dynatrace

MARCH 16, 2023

Monitors signals The first attribute of a good SLO is the ability to monitor the four “golden signals”: latency, traffic, error rates, and resource saturation. Includes the end user While SLOs for back-end silos and processes help eliminate internal issues, back-end-based objectives tell only half the story.

DevOps 207

DevOps Latency Metrics Traffic 207

Dynatrace

MAY 17, 2023

To ensure high standards, it’s essential that your organization establish automated validations in an early phase of the software development process—ideally when code is written. In this case, the four golden signals (latency, traffic, errors, and saturation) are derived from span attributes and DQL metric queries via Dynatrace Grail™.

DevOps 180

DevOps Traffic Latency Best Practices 180

The Netflix TechBlog

MARCH 7, 2024

Since its inception , Metaflow has been designed to provide a human-friendly API for building data and ML (and today AI) applications and deploying them in our production infrastructure frictionlessly. In other cases, it is more convenient to share the results via a low-latency API.

Systems 226

Systems Media Cache Open Source 226

Percona

APRIL 17, 2023

The CFQ works well for many general use cases but lacks latency guarantees. The deadline excels at latency-sensitive use cases ( like databases ), and noop is closer to no schedule at all. 25.84 - Total 21.04 zone_reclaim_mode In some OS versions, the vm.zone_reclaim_mode is enabled. 25.84 - Total 21.04

Best Practices 52

Best Practices Design Tuning Database 52

Randon ASCII

SEPTEMBER 8, 2019

While CPU Usage (Precise) is great for seeing how much CPU time a process is using, and why it is sitting idle, the CPU Usage (Sampled) table is the right tool for figuring out where CPU time is being spent. The next step was to see what RuntimeBroker.exe was doing. I drilled down and quickly found 264 samples hitting in KernelBase.dll!

Latency 79

Latency Cache Programming Operating System 79

Percona

NOVEMBER 3, 2023

About the cluster Following a step-by-step process, the objective is to create a four-node cluster consisting of: PostgreSQL version 15 Citus extension (I’ll be using version 11, but there are newer ones available.) psql pgbench <<_eof1_ qecho adding node citus3.

Database 104

Database Benchmarking Latency C++ 104

Highly Scalable

AUGUST 20, 2013

The shortcomings and drawbacks of batch-oriented data processing were widely recognized by the Big Data community quite a long time ago. It became clear that real-time query processing and in-stream processing is the immediate need in many practical applications. Fault-tolerance.

Big Data 154

Big Data Processing Lambda Database 154

Scalegrid

MARCH 28, 2024

Redis Data Types and Structures The design of Redis’s data structures emphasizes versatility. Redis can be configured to optimally utilize both RDB and AOF persistence methods optimally, achieving a balance between speed and data safety while minimizing the impact on response times due to its child process handling for disk writes.

Cache 130

Cache Storage Scalability Architecture 130

The Netflix TechBlog

JUNE 13, 2023

Replay traffic testing gives us the initial foundation of validation, but as our migration process unfolds, we are met with the need for a carefully controlled migration process. A process that doesn’t just minimize risk, but also facilitates a continuous evaluation of the rollout’s impact.

Traffic 279

Traffic Metrics Systems Strategy 279

Dynatrace

SEPTEMBER 13, 2023

This is a set of best practices and guidelines that help you design and operate reliable, secure, efficient, cost-effective, and sustainable systems in the cloud. This process enables you to continuously evaluate software against predefined quality criteria and service level objectives (SLOs) in pre-production environments.

AWS 240

AWS Efficiency Azure Cloud 240

The Netflix TechBlog

FEBRUARY 18, 2022

Event Prioritization Considering the use cases were wide ranging both in terms of their sources and their importance, we built segmentation into the event processing. We thus assigned a priority to each use case and sharded event traffic by routing to priority-specific queues and the corresponding event processing clusters.

Systems 334

Systems Traffic Architecture Mobile 334

The Netflix TechBlog

DECEMBER 21, 2020

There are several benefits of such optimizations like saving on storage, faster query time, cheaper downstream processing, and an increase in developer productivity by removing additional ETLs written only for query performance improvement. Then deep dive into the merging use case of AutoOptimize and share some results and benefits.

Storage 203

Storage Latency Efficiency Data Engineering 203

Scalegrid

FEBRUARY 8, 2024

Their design emphasizes increasing availability by spreading out files among different nodes or servers — this approach significantly reduces risks associated with losing or corrupting data due to node failure. This process effectively duplicates essential parts of information to safeguard against potential loss.

Storage 130

Storage Systems Big Data Azure 130

The Netflix TechBlog

AUGUST 30, 2021

The challenge, then, is to be able to ingest and process these events in a scalable manner, i.e., scaling with the number of devices, which will be the focus of this blog post. In-Order Processing The semantics of correct device information updates ingestion requires that messages be consumed in the order that they are produced.

Latency 213

Latency Traffic Transportation Hardware 213

Dynatrace

AUGUST 27, 2019

Welcome back to the blog series in which we share how you can easily solve three common problem scenarios by using Dynatrace and xMatters Flow Designer. In Part 1 we explored how DevOps teams can prevent a process crash from taking down services across an organization in five easy steps. xMatters creates and updates Jira issues.

Systems 176

Systems Latency DevOps Azure 176

The Netflix TechBlog

NOVEMBER 17, 2022

A distinct, NN-based, video processing block can evolve independently, be used beyond video downscaling and be combined with different codecs. Our approach to NN-based video downscaling The deep downscaler is a neural network architecture designed to improve the end-to-end video quality by learning a higher-quality video downscaler.

Network 292

Network Media Innovation Efficiency 292

Scalegrid

JANUARY 8, 2024

Key Takeaways AI integration in cloud computing increases operational efficiency by automating processes, optimizing resource allocation, and improving scalability, leading to cost savings and allowing IT teams to concentrate on strategic initiatives. Discover how AI is reshaping the cloud and what this means for the future of technology.

Artificial Intelligence 147

Artificial Intelligence Cloud Scalability Analytics 147

The Netflix TechBlog

AUGUST 29, 2023

Today we have a wealth of tools, both OSS and commercial, all designed for cloud-native environments. For Inter-Process Communication (IPC) between services, we needed the rich feature set that a mid-tier load balancer typically provides. In 2010, however, nearly none of it existed: the CNCF wasn’t formed until 2015!

Traffic 220

Traffic Latency Cloud C++ 220

Dynatrace

JANUARY 31, 2024

The rise of data observability in DevOps Data forms the foundation of decision-making processes in companies across the globe. For DevOps teams that inform deployment strategies, optimize processes, and drive continuous improvement, the integrity and timeliness of data are of significant importance.

DevOps 189

DevOps Analytics Airlines Metrics 189

The Netflix TechBlog

OCTOBER 27, 2020

Usually Data scientists and engineers write Extract-Transform-Load (ETL) jobs and pipelines using big data compute technologies, like Spark or Presto , to process this data and periodically compute key information for a member or a video. The processed data is typically stored as data warehouse tables in AWS S3.

Latency 243

Latency Storage Big Data Tuning 243

Dynatrace

APRIL 5, 2021

AWS Lambda is a serverless compute service that can run code in response to predetermined events or conditions and automatically manage all the computing resources required for those processes. Real-time file processing, for quickly indexing files, processing logs, and validating content.

Lambda 175

Lambda AWS Serverless Hardware 175

The Netflix TechBlog

SEPTEMBER 8, 2020

For each route we migrated, we wanted to make sure we were not introducing any regressions: either in the form of missing (or worse, wrong) data, or by increasing the latency of each endpoint. Being able to canary a new route let us verify latency and error rates were within acceptable limits. This meant that data that was static (e.g.

Latency 233

Latency Cache Java Traffic 233

Dynatrace

FEBRUARY 24, 2021

DevOps is focused on optimizing software development and delivery, and SRE is focused on operations processes. DevOps is not a specific process, but rather a general collection of flexible software creation and delivery practices that looks to close the gap between software development and IT operations. Reduced latency.

DevOps 184

DevOps Software Engineering Speed Google 184

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. Its goal is to assign running processes to time slices of the CPU in a “fair” way. So why mess with it?

Cache 251

Cache Latency Airlines Logistics 251

Dynatrace

JULY 24, 2023

We will also explore the evolution of DevOps automation and the significance of data-driven answers in unlocking streamlined, automated DevOps and SRE processes. Business process automation Business process automation is the foundation for improving operational efficiency.

DevOps 213

DevOps Traffic Efficiency Servers 213

Dynatrace

DECEMBER 15, 2022

This transition to public, private, and hybrid cloud is driving organizations to automate and virtualize IT operations to lower costs and optimize cloud processes and systems. This includes response time, accuracy, speed, throughput, uptime, CPU utilization, and latency. So, what is ITOps? How does ITOps create business value?

Artificial Intelligence 195

Artificial Intelligence DevOps Hardware Virtualization 195