The Netflix TechBlog

DECEMBER 7, 2020

Also notice that this is an ongoing effort, so stay tuned for more speed improvements. For historical reasons, the libvmaf C library has been a minimal solution to integrate VMAF with FFmpeg. Most recently, a tune=vmaf mode was introduced in the libaom library as an option to perform quality-optimized AV1 encoding.

Metrics 159

Metrics Open Source Tuning Speed 159

The Netflix TechBlog

AUGUST 29, 2023

Keeping feature parity between all of these implementations and ensuring that they all behave the same way is challenging: what we want is a single, well-tested implementation of all of this functionality, so we can make changes and fix bugs in one place. This is the first in a series of posts on our journey to service mesh, so stay tuned.

Traffic 229

Traffic Latency Cloud C++ 229

Percona

OCTOBER 4, 2023

As a result, both A and C directly inherit permissions from B and indirectly from D. As a result, both A and C directly inherit permissions from B and indirectly from D. In simpler language, we can view A and C as indirect descendants of D within the role hierarchy. Download Percona Distribution for PostgreSQL Today!

C++ 122

C++ Open Source Tuning Database 122

Percona

MARCH 1, 2023

So you get the Umaluts, Cedils, and C-J-K Language support in your data, plus emojis. EXPLAIN ANALYSE is a big boon to query tuning. Add in INVISIBLE INDEXes, and you can test the efficiency of an index without risking a disastrous rebuild after a delete. The default character set of UTF8MB4 gives Unicode version 9.0

Servers 133

Servers Open Source Tuning C++ 133

The Netflix TechBlog

MARCH 7, 2019

We have to do it at Netflix’s scale: For hundreds of millions of users across hundreds of concurrent tests, spanning many deployment strategies from traditional A/B experiments, to evolving areas like quasi experiments. C) Compression Algorithms should be able to work on raw data as well as compressed data.

Design 175

Design Engineering Innovation Java 175

O'Reilly

FEBRUARY 25, 2025

For some context, over the past 15 years Ive been operating Python Tutor ( [link] ), a free online tool that tens of millions of people around the world have used to write, run, and visually debug their code (first in Python and now also in Java, C, C++, and JavaScript). But in my initial tests a small model like Llama 3.2

Code 81

Code Programming Servers Cloud 81

SQL Performance

DECEMBER 9, 2020

In a comment to my article, Marcos Kirchner mentioned that he tested my solution with varying table value constructor cardinalities, and got different execution times for the different cardinalities. Testing different base table cardinalities is just one dimension to try. Let’s proceed to performance testing.

C++ 114

C++ Servers Testing Processing 114

Percona

MAY 1, 2023

Out of the box, the default PostgreSQL configuration is not tuned for any particular workload. It is primarily the responsibility of the database administrator or developer to tune PostgreSQL according to their system’s workload. What is PostgreSQL performance tuning? Why is PostgreSQL performance tuning important?

Tuning 52

Tuning Database Efficiency Performance 52

Percona

NOVEMBER 9, 2023

With this data, we’ll be able to better understand and tune our Postgres servers. The main idea is to be able to easily detect what the server is doing at any point in time so we can better understand how it reacts to the workload and tune accordingly. You can search the blog for more information on this if needed.

Monitoring 125

Monitoring Tuning Metrics Servers 125

The Morning Paper

OCTOBER 18, 2020

On synthetic datasets containing double-precision keys following a standard normal distribution, the authors compared Learned Sort to a variety of cache-optimized and highly tuned C++ implementations of alternative sorting algorithms, presenting the most competitive alternatives in their results.

Cache 137

Cache C++ Java Tuning 137

Percona

NOVEMBER 3, 2023

Using its default tpc-b benchmark, one can stress test a database of any size ranging from a few clients to simulating thousands interacting with a system sized into the Terabytes if needs be. And now, execute the benchmark: -- execute the following on the coordinator node pgbench -c 20 -j 3 -T 60 -P 3 pgbench The results are not pretty.

Database 114

Database Benchmarking Latency C++ 114

Percona

AUGUST 29, 2023

Depending on the configuration, one can tune a hardware RAID for either performance or redundancy. Now let’s stretch our imagination and consider a second method of high availability, ala Citus. The best way to describe the Citus way of doing things is to reflect how data is managed by a disk RAID array. citus_columnar | 11.1-1

C++ 98

C++ Database Storage Testing 98

The Netflix TechBlog

JULY 1, 2019

In a previous post , we described how our TV application consists of a C++ SDK installed natively on the device, an updatable JavaScript user interface (UI) layer, and a custom rendering layer known as Gibbon. In this case, the test was measuring surface cache after every step.

Cache 167

Cache Design Testing Development 167

HammerDB

JULY 14, 2021

Also, the additional installation of the MySQL client libraries was needed when testing MariaDB. adds full support for MariaDB in both the GUI as shown: And in the CLI as a separate database from MySQL, this post describes a setup and test for MariaDB using the CLI on Linux as an example running the TPROC-C workload.

Testing 58

Testing C++ Virtualization Tuning 58

Brendan Gregg

OCTOBER 8, 2018

C @ns: [256, 512) 10900 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ | [512, 1k) 18291 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@| [1k, 2k) 4998 |@@@@@@@@@@@@@@ | [2k, 4k) 57 | | [4k, 8k) 117 | | [8k, 16k) 48 | | [16k, 32k) 109 | | [32k, 64k) 3 | |. Attaching 2 probes. ^C eBPF does more. that uses common functionality.

C++ 105

C++ Virtualization Programming Latency 105

HammerDB

MAY 5, 2019

Fortunately the HammerDB TPC-C/OLTP workload intentionally has a great deal of contention between threads and is therefore ideal for testing spin-locks. So to test I took a system with Skylake CPUs and all storage on a P4800X SSD. In the recent MySQL 8.0.16 model name : Intel(R) Xeon(R) Platinum 8180 CPU @ 2.50GHz. linux-glibc2.12-x86_64/data

Testing 49

Testing Tuning Storage Latency 49

Percona

OCTOBER 17, 2023

Many changes and new features are brought to the system, and as part of keeping in tune with the changes and how they can impact us, we go through the changes to better understand them. As almost a tradition, a newer major release of MongoDB is born every year, and this year, it’s no different. From that, this article is born.

Metrics 120

Metrics C++ Database Storage 120

Smashing Magazine

AUGUST 9, 2021

This explains the challenges involved in deploying and testing HTTP/3 yourself. For example, if we have files A, B, and C, we would have three TCP connections. You see, because TCP is a much older protocol and not made for just loading web pages, it doesn’t know about A, B, or C. This is more in depth and technical.

Transportation 145

Transportation Internet Internet Network 145

The Netflix TechBlog

OCTOBER 18, 2019

Contextual and Multi-armed Bandits enable faster and adaptive alternatives to traditional A/B Testing. Some simulation results and some online A/B test results were shared, demonstrating substantial lifts in the primary business metrics, while keeping the secondary metrics above preset guardrails.

Infrastructure 131

Infrastructure Metrics Architecture Efficiency 131

Brendan Gregg

SEPTEMBER 25, 2021

As (C) looked like a kernel rebuild, I started with (D) and (E). ## 5. There's also a test and println() in the loop to, hopefully, convince the compiler not to optimize-out an otherwise empty loop. This will slow this test a little.) I also rewrote this in C and called gettimeofday(2) directly: $ cat gettimeofdaybench.c.

Speed 103

Speed Java AWS Virtualization 103

O'Reilly

SEPTEMBER 19, 2023

You can download these models to use out of the box, or employ minimal compute resources to fine-tune them for your particular task. You can see a simulation as a temporary, synthetic environment in which to test an idea. Millions of tests, across as many parameters as will fit on the hardware. Specifically, through simulation.

Hardware 135

Hardware Storage Big Data Blockchain 135

The Morning Paper

JUNE 2, 2019

Developing machine learning models is no different from developing traditional software, in the sense that it is also a full life cycle involving design, implementation, tuning, testing, and deployment. So test sets used for CI have a ‘shelf-life’, after which they need to be refreshed. ci will raise an alarm (e.g.

C++ 60

C++ Strategy Testing Servers 60

HammerDB

FEBRUARY 25, 2021

HammerDB is a load testing and benchmarking application for relational databases. All the databases that HammerDB tests implement a form of MVCC (multi-version concurrency control). For HammerDB both TPROC-C and TPROC-H run all of their workloads on the database being tested in SQL. Database interfaces in C.

Programming 68

Programming Benchmarking Virtualization C++ 68

Percona

AUGUST 30, 2023

Those parameters taken from the specific data directory initialization, which overrides the built-in parameters, can be checked like this: select name,setting from pg_settings where source='override'; This override includes some of the calculated auto-tune values for that environment. Type "help" for help.

Database 120

Database Servers Open Source Tuning 120

Smashing Magazine

APRIL 13, 2020

On top of basic font style properties, we can now reliably fine-tune OpenType figures, hyphenation, ligatures, and even kerning. But if you’re still concerned about supporting a legacy browser, consider using a feature query to test for text-stroke support and provide an appropriate fallback for them. Large preview ).

Design 132

Design Media C++ Technology 132

SQL Performance

DECEMBER 9, 2020

In a comment to my article, Marcos Kirchner mentioned that he tested my solution with varying table value constructor cardinalities, and got different execution times for the different cardinalities. Testing different base table cardinalities is just one dimension to try. Let’s proceed to performance testing.

C++ 61

C++ Servers Testing Processing 61

HammerDB

APRIL 19, 2023

Therefore, before we attempt to measure our database performance, we should know the system or cloud instance to be tested in detail. This all sounded very similar to HammerDB TPROC-C workload (we will look at HammerDB TPROC-H (OLAP) another time), so it is easy for any reader to think both workloads are about the same. 4.22 %usr 38.40

Performance 55

Performance Benchmarking Cache Storage 55

HammerDB

OCTOBER 18, 2019

Well its difficult to be entirely sure however the tests have all the characteristics of tests observed previously where the CPUs are running in powersave mode. So lets take an Ubuntu system with Platinum 8280 CPUs with the following Ubuntu OS, reboot and check the CPU configuration before running any tests. The answer is no.

Benchmarking 63

Benchmarking Performance Hardware C++ 63

Rigor

JANUARY 13, 2020

Steve Souders, plus Dion Almaer, Doug Crockford, Ben Galbraith, Tony Gentilcore, Dylan Schiemann, Stoyan Stefanov, Nicole Sullivan, Nicholas C. This practical guide shows users how run tests & interpret results to gain a better and more thorough understanding of hidden features in WebPageTest. Web Performance Tuning.

Performance 75

Performance Social Media Website Website Performance 75

Brendan Gregg

FEBRUARY 9, 2018

In this post I'll look at the Linux kernel page table isolation (KPTI) patches that workaround Meltdown: what overheads to expect, and ways to tune them. Much of my testing was on Linux 4.14.11 Also microbenchmarks, which often stress-test the system, will suffer the largest losses. and 4.14.12 DTLB and 4.4% DTLB and 1.1%

Performance 72

Performance Benchmarking Cache Tuning 72

Smashing Magazine

SEPTEMBER 6, 2021

This explains the challenges involved in deploying and testing HTTP/3 yourself. In our own early tests , I found seriously diminishing returns at about 40 files. As such, a micro-optimization is, again, how you probably need to fine-tune things on a low level to really benefit from it. This is more in-depth and technical.

Network 120

Network Servers Cache Traffic 120

The Netflix TechBlog

OCTOBER 18, 2019

Contextual and Multi-armed Bandits enable faster and adaptive alternatives to traditional A/B Testing. Some simulation results and some online A/B test results were shared, demonstrating substantial lifts in the primary business metrics, while keeping the secondary metrics above preset guardrails.

Infrastructure 100

Infrastructure Metrics Architecture Efficiency 100

Abhishek Tiwari

SEPTEMBER 15, 2024

million lines of C code. The choice of C as the programming language is telling. The choice of C as the programming language is telling. This allows NASA's engineers to fine-tune every aspect of the rover's behavior, optimizing for reliability rather than rapid development or ease of use.

Engineering 52

Engineering Social Media Hardware C++ 52

Percona

JANUARY 17, 2019

Test data and versions. For my test, I need to choose: Aurora instance type and comparison. For the purposes of the test I turned PQ on and off (normally AWS Aurora uses its own heuristics to determine if the PQ will be helpful or not): Turn on and force: mysql> set session aurora_pq = 1; Query OK, 0 rows affected (0.00

Cache 71

Cache C++ AWS Airlines 71

HammerDB

OCTOBER 18, 2018

An essential part of database performance testing is viewing the statistics generated by the database during the test and in 2009 HammerDB introduced automatic AWR snapshot generation for Oracle for the TPC-C test. However what if you want to review performance data in real time as the test is running?

Benchmarking 40

Benchmarking Monitoring Performance C++ 40

Brendan Gregg

SEPTEMBER 5, 2017

We're now testing container hosts in production on btrfs, instead of ZFS. Hit Ctrl-C to end. ^C C operation = 'read' usecs : count distribution 0 -> 1 : 0 | | 2 -> 3 : 0 | | 4 -> 7 : 4479 | *| 8 -> 15 : 1028 | | 16 -> 31 : 14 | | 32 -> 63 : 1 | | [.]. Tracing ZFS operation latency.

Virtualization 75

Virtualization AWS Engineering Hardware 75

Percona

SEPTEMBER 13, 2018

For our testing we decided to capture all the SELECT queries that were hitting our Amazon Aurora instance, mainly because we had a sysbench OLTP read only workload and that wouldn’t really have a lot of slow queries. On our Aurora test slow log file, the initial results didn’t look right so we had to apply a workaround.

C++ 60

C++ AWS Open Source Testing 60