The Netflix TechBlog

JUNE 13, 2023

By collecting and analyzing key performance metrics of the service over time, we can assess the impact of the new changes and determine if they meet the availability, latency, and performance requirements. They enable us to further fine-tune and configure the system, ensuring the new changes are integrated smoothly and seamlessly.

Traffic 279

Traffic Metrics Systems Strategy 279

Scalegrid

JUNE 22, 2020

We are going to use a common, popular plan size using the below configurations for this performance benchmark: Comparison Overview. MySQL Configuration Management & Tuning. Business support only available for accounts over $500/month. DigitalOcean. Instance Type. Medium: 4 vCPUs. Medium: 4 vCPUs. MySQL Version. Standalone.

Database 217

Database Benchmarking Latency Performance 217

Smashing Magazine

SEPTEMBER 29, 2021

Tip: When evaluating quality, compression and fine-tuning of modern formats, Squoosh.app ’s ability to perform a visual side-by-side comparison is helpful. Here, you can see a size comparison between a JPEG image and its corresponding (lossy) AVIF image converted using the Squoosh app: ( Large preview ). Large preview ).

Open Source 144

Open Source Speed Website Google 144

The Netflix TechBlog

FEBRUARY 13, 2024

The remainder of this post discusses the first paper above, which was published at KDD ’22 (and available on ArXiV ). These two plots look pretty close, but we can do better than an eyeball comparison — and we want the computer to be able to continuously evaluate if there is any significant difference between the distributions.

Testing 239

Testing Software Software Metrics 239

Dynatrace

JUNE 8, 2021

Observability is a topic at the top of mind for all architects, Site Reliability Engineers (SREs), and more – each wanting to use observability to proactively detect issues and guarantee the best experience and availability to users. Metric 2 (Number of requests in error): It’s already available in the Dynatrace built-in metric.

Metrics 170

Metrics Tuning Programming Systems 170

The Netflix TechBlog

JUNE 15, 2022

This post will discuss how Studio Search supports querying the data available in these indices. The DSL supports specifying the search criteria as comparison expressions or inclusion/exclusion filters. Through the power of federation, users are then able to hydrate the search results with any data available in the federated graph.

Engineering 183

Engineering Architecture Availability Tuning 183

SQL Performance

JANUARY 23, 2020

Tom Davidson, Opening Microsoft's Performance-Tuning Toolbox SQL Server Pro Magazine, December 2003. Waits and Queues has been used as a SQL Server performance tuning methodology since Tom Davidson published the above article as well as the well-known SQL Server 2005 Waits and Queues whitepaper in 2006. The Top Queries That Weren't.

Tuning 115

Tuning Servers Monitoring Hardware 115

Dynatrace

MAY 9, 2019

A comparison of the previous and new steps to accomplish the same goals is shown in the table below. exe files is available with OneAgent version 1.167 and Dynatrace SaaS version 1.168. Pretty easy, right? “How does the new process compare to the old one exactly?” Differences are marked in bold. Previous solution.

Storage 116

Storage Tuning Traffic Architecture 116

The Netflix TechBlog

NOVEMBER 2, 2021

Perceptual quality measurements are used to drive video encoding optimizations , perform video codec comparisons , carry out A/B testing and optimize streaming QoE decisions to mention a few. Here, based on the video length, the throughput and latency requirements, available scale etc., The workflow is initiated. 4c & 5.

Media 171

Media Innovation Metrics Latency 171

Dynatrace

OCTOBER 28, 2019

Here too, you can use charts to visualize data comparisons between two timeframes. The enhancements to user sessions queries are available with the release of Dynatrace version 1.178. They’re available for all users by default. Stay tuned! Are you as excited as we are? We bet you are! It doesn’t stop here!

Website 103

Website Metrics Tuning Analytics 103

All Things Distributed

MARCH 19, 2018

Post-training model tuning and rich states. Processing massively scalable datasets in a streaming manner poses a challenge for model tuning, also known as hyperparameter optimization (HPO). When a comparison is possible, Amazon SageMaker algorithms often run more than 10x faster than other ML solutions like Spark ML.

Scalability 159

Scalability Hardware AWS Tuning 159

All Things Distributed

NOVEMBER 30, 2016

Effectively applying AI involves extensive manual effort to develop and tune many different types of machine learning and deep learning algorithms (e.g. automatic speech recognition, natural language understanding, image classification), collect and clean the training data, and train and tune the machine learning models. Amazon Lex.

AWS 165

AWS Lambda Artificial Intelligence Mobile 165

Percona

DECEMBER 20, 2018

I found the comparison of InnoDB vs. MyISAM quite interesting, and I’ll use it in this post. Linux OS Tuning for MySQL Database Performance. In this post we will review the most important Linux settings to adjust for performance tuning and optimization of a MySQL database server. Tuning InnoDB Primary Keys.

Database 40

Database Tuning Performance AWS 40

Dynatrace

JULY 23, 2019

All the sample code including instrumentation is available on GitHub. The OneAgent SDK is available as a package called oneagent-sdk in the PyPI repository. For example, you can create a comparison of response times when publishing a new blog post split by user role. Let’s walk through how. Tracing database requests.

Monitoring 136

Monitoring Software Software Database 136

The Morning Paper

OCTOBER 8, 2019

The core application of machine learning discussed in this paper is the model which orders available listings according to a guest’s likelihood of booking. Note that the Airbnb team already had experience with machine learning, data pipelines in place, and an online controlled experiment platform available. Tuning the model.

Network 82

Network Architecture Tuning Traffic 82

Brendan Gregg

SEPTEMBER 25, 2021

As I'm interested in the relative comparison I can just compare the total runtimes (the "real" time) for the same result. Checking those available: $ cat /sys/devices/system/clocksource/clocksource0/available_clocksource. How long is each time call? Assuming the loop is dominated by the time call, it works out to be about 0.13

Speed 126

Speed Java AWS Virtualization 126

Smashing Magazine

APRIL 11, 2022

However, developers with a deep understanding of the project may want to improve performance beyond that by doing some fine-tuning under the hood. We can run Lighthouse test to check the metrics and use this data for comparison. Fine-tuning JavaScript Resource Priorities. Without priority hints. Large preview ).

Media 70

Media Tuning Metrics Google 70

Percona

JANUARY 19, 2023

sec) Records: 0 Duplicates: 0 Warnings: 0 mysql> INSERT INTO employees_compressed SELECT * FROM employees; Size comparison: [user1] percona@db1: ~ $ sudo ls -lh /var/lib/mysql/employees/|grep employees -rw-r --. It is available under a paid subscription. 1 mysql mysql 704M Dec 30 02:28 employees.ibd -rw-r --.

Open Source 86

Open Source Storage Database Big Data 86

The Morning Paper

JUNE 4, 2019

To detect feature skew, the validate does a key-join between corresponding batches of training and serving data followed by a feature-wise comparison. There are classic distance measures such as KL-divergence and cosine similarity, but product teams had a hard time understanding what they really meant and hence how to tune thresholds.

Google 92

Google Code Infrastructure Best Practices 92

Speed Curve

DECEMBER 25, 2013

Source: Guy Podjarny However, we do now have a full set of techniques to effectively deliver highly performative sites that not only visually scale across devices but also deliver code and assets tuned to the width of a device. Comparison of page size and assets types across different responsive widths. Source: SpeedCurve.

Performance 60

Performance Mobile Website Benchmarking 60

Brendan Gregg

SEPTEMBER 25, 2021

As I'm interested in the relative comparison I can just compare the total runtimes (the "real" time) for the same result. Trying it out: centos$ time java TimeBench real 0m12.989s user 0m3.368s sys 0m18.561s ubuntu# time java TimeBench real 1m8.300s user 0m38.337s sys 0m29.875s How long is each time call? us on Centos and 0.68 us on Ubuntu.

Speed 40

Speed Java AWS Virtualization 40

Smashing Magazine

SEPTEMBER 6, 2021

As such, a micro-optimization is, again, how you probably need to fine-tune things on a low level to really benefit from it. I would personally recommend using Cloudflare or Fastly because they have been intimately involved in the standardization process and will have the most advanced and well-tuned implementations available.

Network 105

Network Servers Cache Traffic 105

Smashing Magazine

AUGUST 22, 2021

Additionally, even if we knew the available physical bandwidth, that wouldn’t mean we could use all of it ourselves. Several users are typically active on a network concurrently, each of whom need a fair share of the available bandwidth. More in-depth information is available. What Does It All Mean?

Performance 106

Performance Network Latency Servers 106

John McCalpin

JANUARY 22, 2018

Since “peak performance” for the processor is two (512-bit SIMD) FMA instructions per cycle, any instructions that are not FMA instructions subtract directly from the available peak performance. of the “adjusted peak performance”, there is no longer a significant upside to performance tuning. 8.056 0.056 75.0%

Latency 40

Latency Hardware Code Testing 40

Smashing Magazine

AUGUST 9, 2021

HTTP/2 versus HTTP/3 protocol stack comparison ( Large preview ). ISPs and intermediate networks might block it because metrics such as average delays and packet loss percentages are no longer easily available, making it more difficult to detect and diagnose problems. Stay tuned! (vf, QUIC has a higher encryption overhead.

Transportation 145

Transportation Internet Internet Network 145

HammerDB

OCTOBER 12, 2018

Finally it is also important to note that this comparison is focused around OLTP based workloads, HammerDB also supports a TPC-H based workload for analytics with complex ad-hoc queries. GHz available cpufreq governors: performance powersave . CPUs which need to have their frequency coordinated by software: 0 . innodb_file_per_table.

Best Practices 40

Best Practices Scalability Performance C++ 40

Brendan Gregg

SEPTEMBER 5, 2017

What follows are topics that may be of interest to anyone looking to migrate their systems and skillset: scan these to find topics that interest you. ## ZFS ZFS is available for Linux via the [zfsonlinux] and [OpenZFS] projects, and more recently was included in Canonical's Ubuntu Linux distribution: Ubuntu Xenial 16.04 LTS (April 2016).

Virtualization 76

Virtualization AWS Engineering Hardware 76

Percona

JANUARY 17, 2019

On multi-core machines – which is the majority of the hardware nowadays – and in the cloud, we have multiple cores available for use. For my test, I need to choose: Aurora instance type and comparison. Aurora instance type and comparison. With faster disks (i.e. AWS Aurora (based on MySQL 5.6) Test data and versions.

Cache 46

Cache C++ AWS Airlines 46

SQL Server According to Bob

JANUARY 11, 2020

​​ Various defragmentation ​​ utilities ​​ are ​​ available ​​ on the market today. Therefore, the database can have pages in any one of the three available verification states. SQL Server uses read-ahead logic to avoid query stalls caused by I/O waits.

Servers 40

Servers Cache Database Media 40