The Netflix TechBlog

FEBRUARY 14, 2025

This dual-path approach leverages Kafkas capability for low-latency streaming and Icebergs efficient management of large-scale, immutable datasets, ensuring both real-time responsiveness and comprehensive historical data availability. This integration will not only optimize performance but also ensure more efficient resource utilization.

Tuning 166

Tuning Latency Efficiency Storage 166

Dynatrace

JANUARY 15, 2020

This extends Dynatrace visibility into Citrix user experience and Citrix platform performance. Therefore, it requires multidimensional and multidisciplinary monitoring: Infrastructure health —automatically monitor the compute, storage, and network resources available to the Citrix system to ensure a stable platform. Citrix VDA.

Latency 52

Latency Performance Virtualization Infrastructure 52

The Netflix TechBlog

SEPTEMBER 18, 2024

Firstly, developers struggled to reason about consistency, durability and performance in this complex global deployment across multiple stores. These include challenges with tail latency and idempotency, managing “wide” partitions with many rows, handling single large “fat” columns, and slow response pagination.

Latency 260

Latency Storage Cache Servers 260

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. Our object storage service splits objects into many parts and stores them in S3.

Media 218

Media Storage Processing Cache 218

DZone

AUGUST 16, 2023

Caching is the process of storing frequently accessed data or resources in a temporary storage location, such as memory or disk, to improve retrieval speed and reduce the need for repetitive processing.

Cache 246

Cache Scalability Performance Latency 246

Dynatrace

DECEMBER 13, 2023

Mobile applications (apps) are an increasingly important channel for reaching customers, but the distributed nature of mobile app platforms and delivery networks can cause performance problems that leave users frustrated, or worse, turning to competitors. What is mobile app performance? Issue remediation.

Best Practices 246

Best Practices Mobile Metrics Performance 246

Dynatrace

OCTOBER 23, 2023

Secondly, determining the correct allocation of resources (CPU, memory, storage) to each virtual machine to ensure optimal performance without over-provisioning can be difficult. This presents a challenge for IT operations teams, specifically in identifying and addressing performance issues or planning how to prevent future issues.

Efficiency 264

Efficiency Virtualization Hardware Performance 264

Scalegrid

APRIL 28, 2020

Since database hosting is more dependent on memory (RAM) than storage, we are going to compare various instance sizes ranging from just 1GB of RAM up to 64GB of RAM so you can see how costs vary across different application workloads. High performance. Does it affect latency? Yes, you can see an increase in latency.

Azure 344

Azure AWS Database Latency 344

Dynatrace

OCTOBER 3, 2024

Dynatrace OTel Collector Understand your applications with ease Due to a lack of contextual insights and actionable intelligence, application teams often find themselves overwhelmed by data, unable to quickly identify the root causes of performance issues. There is no need to think about schema and indexes, re-hydration, or hot/cold storage.

Performance 290

Performance Architecture Innovation Latency 290

The Netflix TechBlog

DECEMBER 17, 2024

The Challenge of Title Launch Observability As engineers, were wired to track system metrics like error rates, latencies, and CPU utilizationbut what about metrics that matter to a titlessuccess? By logging all titles as they are displayed, we can process these logs to identify anomalies and gain insights into system performance.

Traffic 172

Traffic Scalability Strategy Monitoring 172

High Scalability

JANUARY 11, 2022

from a client it performs two parallel operations: i) persisting the action in the data store ii) publish the action in a streaming data store for a pub-sub model. User Feed Service, Media Counter Service) read the actions from the streaming data store and performs their specific tasks. After that, the various services (e.g.

Design 334

Design Media Storage Logistics 334

The Netflix TechBlog

OCTOBER 8, 2024

Rajiv Shringi Vinay Chella Kaidan Fullerton Oleksii Tkachuk Joey Lynch Introduction As Netflix continues to expand and diversify into various sectors like Video on Demand and Gaming , the ability to ingest and store vast amounts of temporal data — often reaching petabytes — with millisecond access latency has become increasingly vital.

Latency 239

Latency Storage Traffic Tuning 239

DZone

NOVEMBER 25, 2023

We often dwell on the technical aspects of database selection, focusing on performance metrics , storage capacity, and querying capabilities. The New Decision Matrix: Beyond Performance Metrics Performance metrics are pivotal, no doubt. Yet, they are often evaluated in isolation, removed from the business context.

Database 278

Database Latency Speed Metrics 278

High Scalability

SEPTEMBER 8, 2018

You get all the multicore Anna performance you want, but you don’t pay for what you don’t need. Just to throw out some numbers, we measured Anna providing 355x the performance of DynamoDB for the dollar. Each storage server collects statistics about the requests it serves, the data it stores, etc.

Storage 199

Storage Performance AWS Media 199

Scalegrid

MAY 11, 2024

MongoDB offers several storage engines that cater to various use cases. The default storage engine in earlier versions was MMAPv1, which utilized memory-mapped files and document-level locking. This allowed for sequential access and indexed access, but random writes could cause performance issues.

Storage 130

Storage Engineering Cache Database 130

The Netflix TechBlog

SEPTEMBER 24, 2021

The ProRes codec family provides great editing performance and image quality. From chunk encoding to assembly and packaging, the result of each previous processing step must be uploaded to cloud storage and then downloaded by the next processing step. Uploading and downloading data always come with a penalty, namely latency.

Cloud 242

Cloud Media Storage Cache 242

Scalegrid

FEBRUARY 24, 2025

Scaling RabbitMQ ensures your system can handle growing traffic and maintain high performance. Optimizing RabbitMQ performance through strategies such as keeping queues short, enabling lazy queues, and monitoring health checks is essential for maintaining system efficiency and effectively managing high traffic loads.

Best Practices 147

Best Practices Traffic Strategy Scalability 147

DZone

JUNE 15, 2023

In the realm of production applications, it is customary to address performance bottlenecks such as CPU and memory issues in order to identify and resolve their underlying causes. It is worth noting that this data collection process does not impact the performance of the application.

Latency 253

Latency Storage Strategy Metrics 253

The Netflix TechBlog

MAY 4, 2023

The first phase involves validating functional correctness, scalability, and performance concerns and ensuring the new systems’ resilience before the migration. These include Quality-of-Experience(QoE) measurements at the customer device level, Service-Level-Agreements (SLAs), and business-level Key-Performance-Indicators(KPIs).

Traffic 347

Traffic Latency Tuning Systems 347

Dynatrace

SEPTEMBER 23, 2020

Storage mount points in a system might be larger or smaller, local or remote, with high or low latency, and various speeds. Sometimes these locations landed on mount points which, due to capacity, availability, or access constraints, weren’t well suited for large runtime storage. See details below. See details below.

Storage 227

Storage Latency Operating System Network 227

Scalegrid

FEBRUARY 8, 2024

A distributed storage system is foundational in today’s data-driven landscape, ensuring data spread over multiple servers is reliable, accessible, and manageable. Understanding distributed storage is imperative as data volumes and the need for robust storage solutions rise.

Storage 130

Storage Systems Big Data Azure 130

Dynatrace

JANUARY 17, 2024

An AI observability strategy—which monitors IT system performance and costs—may help organizations achieve that balance. AI requires more compute and storage. Training AI data is resource-intensive and costly, again, because of increased computational and storage requirements. AI performs frequent data transfers.

Strategy 288

Strategy Artificial Intelligence Storage Cloud 288

Dynatrace

NOVEMBER 9, 2023

Nutanix overview dashboard The extension automatically gathers real-time performance data from your Nutanix clusters to monitor resource usage, cluster health, and more, all in one place. Dynatrace, in tandem with the Nutanix extension, simplifies performance monitoring and makes issue identification and resolution more efficient.

Virtualization 246

Virtualization Storage Metrics Monitoring 246

Dynatrace

OCTOBER 4, 2022

Data warehouses offer a single storage repository for structured data and provide a source of truth for organizations. Unlike data warehouses, however, data is not transformed before landing in storage. A data lakehouse provides a cost-effective storage layer for both structured and unstructured data. Query language.

Artificial Intelligence 246

Artificial Intelligence Storage Analytics Government 246

Dynatrace

AUGUST 21, 2024

These releases often assumed ideal conditions such as zero latency, infinite bandwidth, and no network loss, as highlighted in Peter Deutsch’s eight fallacies of distributed systems. With Dynatrace, teams can seamlessly monitor the entire system, including network switches, database storage, and third-party dependencies.

Engineering 262

Engineering Systems Latency Metrics 262

Dynatrace

SEPTEMBER 25, 2019

Therefore, it requires multidimensional and multidisciplinary monitoring: Infrastructure health —automatically monitor the compute, storage, and network resources available to the Citrix system to ensure a stable platform. Platform performance —get visibility into the performance of the Citrix platform to optimize application delivery.

Latency 164

Latency Performance Virtualization Infrastructure 164

The Netflix TechBlog

OCTOBER 27, 2020

Iceberg is widely adopted in Netflix as a data warehouse table format that addresses many of the usability and performance problems with Hive tables. The data warehouse is not designed to serve point requests from microservices with low latency. As most key-value storage engines support efficiently deleting a namespace (e.g.

Latency 252

Latency Storage Big Data Tuning 252

The Netflix TechBlog

OCTOBER 19, 2020

If we had an ID for each streaming session then distributed tracing could easily reconstruct session failure by providing service topology, retry and error tags, and latency measurements for all service calls. Our distributed tracing infrastructure is grouped into three sections: tracer library instrumentation, stream processing, and storage.

Infrastructure 238

Infrastructure Transportation Storage Open Source 238

Dynatrace

JUNE 25, 2020

The network latency between cluster nodes should be around 10 ms or less. – A Dynatrace customer, Head of Performance Engineering. For Premium HA, this has been extended from 10 ms latency (in the same network region) to around 100 ms network latency due to asynchronous data replication between regions.

Availability 264

Availability Hardware Latency Traffic 264

The Netflix TechBlog

APRIL 27, 2023

While Atlas is architected around compute & storage separation, and we could theoretically just scale the query layer to meet the increased query demand, every query, regardless of its type, has a data component that needs to be pushed down to the storage layer.

Storage 297

Storage Cache Metrics Database 297

Dynatrace

JUNE 1, 2023

Service level objectives (SLOs) provide a powerful framework for measuring and maintaining software performance, reliability, and user satisfaction. SLOs are a valuable tool for organizations to ensure the health and performance of their applications. Note : you might hear the term latency used instead of response time.

Website 245

Website Latency Traffic Virtualization 245

Percona

MARCH 23, 2023

This blog is in reference to our previous ones for ‘Innodb Performance Optimizations Basics’ 2007 and 2013. Although there have been many blogs about adjusting MySQL variables for better performance since then, I think this topic deserves a blog update since the last update was a decade ago, and MySQL 5.7

Performance 144

Performance Hardware Tuning Storage 144

Dynatrace

SEPTEMBER 13, 2023

The framework comprises six pillars: Operational Excellence, Security, Reliability, Performance Efficiency, Cost Optimization, and Sustainability. And how can you verify this performance consistently across a multicloud environment that also uses Microsoft Azure and Google Cloud Platform frameworks?

AWS 305

AWS Efficiency Azure Cloud 305

The Netflix TechBlog

SEPTEMBER 10, 2024

Finally, the device receives the message, and the action, such as “Show me Stranger Things on Netflix”, is performed. Dynomite had great performance, but it required manual scaling as the system grew. KeyValue is an abstraction over the storage engine itself, which allows us to choose the best storage engine that meets our SLO needs.

Latency 234

Latency Cache Tuning Efficiency 234

The Morning Paper

JANUARY 28, 2020

Narrowing the gap between serverless and its state with storage functions , Zhang et al., Shredder is " a low-latency multi-tenant cloud store that allows small units of computation to be performed directly within storage nodes. " High performance. SoCC’19. "Narrowing Introducing Shredder.

Serverless 99

Serverless Storage Latency Cloud 99

Dynatrace

OCTOBER 2, 2024

There is no need to think about schema and indexes, re-hydration, or hot/cold storage. OpenPipeline ensures data security and privacy—data is collected and processed securely and compliantly, with high-performance filtering, masking, routing, and encryption—and contextualizes incoming data in real time. Figure 11.

Analytics 304

Analytics AWS DevOps Cloud 304

Brendan Gregg

JULY 4, 2021

It's an exciting time for developments in computer performance, not just for the BPF technology (which I often [write about]) but also for processors with 3D stacking and cloud vendor CPUs (e.g., This was a chance to talk about other things I've been working on, such as the present and future of hardware performance.

Performance 121

Performance Latency Hardware Storage 121