Brendan Gregg

AUGUST 29, 2021

1072-aws (xxx) 12/18/2018 _x86_64_ (16 CPU) avg-cpu: %user %nice %system %iowait %steal %idle 5.03 avg-cpu: %user %nice %system %iowait %steal %idle 14.81 avg-cpu: %user %nice %system %iowait %steal %idle 14.86 avg-cpu: %user %nice %system %iowait %steal %idle 14.95 avg-cpu: %user %nice %system %iowait %steal %idle 14.54

Cache 105

Cache C++ AWS Latency 105

Brendan Gregg

DECEMBER 31, 2017

My last talk for 2017 was at AWS re:Invent, on "How Netflix Tunes EC2 Instances for Performance," an updated version of my [2014] talk. A video of the talk is on youtube : The slides are on slideshare : I love this talk as I get to share more about what the Performance and Operating Systems team at Netflix does, rather than just my work.

Tuning 61

Tuning AWS Best Practices Network 61

Brendan Gregg

DECEMBER 30, 2017

My last talk for 2017 was at AWS re:Invent, on "How Netflix Tunes EC2 Instances for Performance," an updated version of my [2014] talk. A video of the talk is on youtube : The slides are on slideshare : I love this talk as I get to share more about what the Performance and Operating Systems team at Netflix does, rather than just my work.

Tuning 52

Tuning AWS Best Practices Network 52

The Morning Paper

JANUARY 21, 2020

In the next post we’ll look at the detailed analysis of how a team at Etsy handled a particular incident on December 4th 2014, to see what we can learn from it. The Internet itself, over which these systems operate, is a dynamically distributed network spanning national borders and policies with no central coordinating agent.

Internet 98

Internet Internet Strategy Energy 98

SQL Performance

FEBRUARY 22, 2021

The system table sys.sysprocesses was replaced way back in SQL Server 2005 by a set of dynamic management views (DMVs), most notably sys.dm_exec_requests , sys.dm_exec_sessions , and sys.dm_exec_connections. We recommend that you use the current SQL Server system views instead. sysprocesses. Why this is a problem.

Latency 57

Latency Servers Monitoring Systems 57

O'Reilly

APRIL 23, 2024

As the internet grew, the amount of information available to consumers became so vast that it outran traditional human means of curation and selection. But in 2013 and 2014, it remained stuck at 83% , and while in the ten years since, it has reached 95% , it had become clear that the easy money that came from acquiring more users was ending.

Social Media 108

Social Media Google Internet Internet 108

ACM Sigarch

APRIL 26, 2018

The new DGX-2, which achieves 2 petaFLOPS in the system, combines 16 fully interconnected GPUS with 10x the deep learning performance. In 2014, Google’s Project Tango deployed Myriad1 to construct 3D indoor maps. Comprehensive lists are available here and here. . The new GV100 packs 7.4 TFLOPS FP-64, 14.8 TFLOPS FP-32 and 118.5

Hardware 99

Hardware Artificial Intelligence Architecture Google 99

Percona

JUNE 4, 2023

Before we dive into the differences between MariaDB and MySQL, we will provide a thorough examination of each relational database management system (RDBMS). While originally designed to be a drop-in replacement for MySQL, it evolved into its own distinct database management system and is now maintained and supported by the MariaDB Foundation.

Open Source 52

Open Source Servers Storage Scalability 52

Brendan Gregg

JULY 2, 2021

This is also applicable for people adding it to their own in-house monitoring systems. You likely already have agents running on all your customer systems. There are so many options it's really your own preference based on your existing system and customer environments. cachestat File system cache statistics line charts 7.

Open Source 40

Open Source Latency Cache Energy 40

VoltDB

JANUARY 12, 2018

Our co-founder, 2014 Turing Award winner, Mike Stonebraker , has a saying: “ Eventual consistency means ‘creates garbage.’” ” One of the suggested benefits of eventual consistency systems is that they support high availability across multiple data centers.But the biggest downside ? No More Eventual Consistency.

Database 40

Database Blockchain Healthcare Transportation 40

VoltDB

JANUARY 12, 2018

Our co-founder, 2014 Turing Award winner, Mike Stonebraker , has a saying: “ Eventual consistency means ‘creates garbage.’” ” One of the suggested benefits of eventual consistency systems is that they support high availability across multiple data centers.But the biggest downside ? No More Eventual Consistency.

Database 40

Database Blockchain Healthcare Transportation 40

Brendan Gregg

AUGUST 29, 2021

1072-aws (xxx) 12/18/2018 _x86_64_ (16 CPU) avg-cpu: %user %nice %system %iowait %steal %idle 5.03 avg-cpu: %user %nice %system %iowait %steal %idle 14.81 avg-cpu: %user %nice %system %iowait %steal %idle 14.86 avg-cpu: %user %nice %system %iowait %steal %idle 14.95 avg-cpu: %user %nice %system %iowait %steal %idle 14.54

Cache 40

Cache C++ AWS Systems 40

Smashing Magazine

DECEMBER 2, 2021

The project is available in twenty languages, and encourages all kinds of contributions to open-source projects — including non-pull-request contributions. Lean UXMas has been publishing each advent since 2014 and is a collection of the most popular articles from this year’s Agile and Lean UX latest news. 24 Pull Requests.

Design 107

Design Development Programming Website 107

Alex Russell

JUNE 22, 2022

There's no point in recommending a better browser because none is available. The New York Times reported in late 2020 that Google now pays Apple between $8-12 billion per year , up from $1 billion in 2014. In 2014, Apple would have enjoyed a profit margin of 50% if it had spent half a billion on browser engineering. Cui bono? —

Engineering 110

Engineering Traffic Government Google 110

Brendan Gregg

MARCH 16, 2024

The problem is that this system has a default libc that has been compiled without frame pointers, so any stack walking stops at the libc layer, producing a partial stack that's missing the application frames. This is pretty common and usually goes unnoticed as the flame graph looks ok at first glance. This can be done using eBPF as well.

Java 145

Java Cache Google Hardware 145

SQL Performance

MARCH 17, 2020

Back in 2014, I wrote an article called Performance Tuning the Whole Query Plan. It has 32GB RAM, with 24GB available to the SQL Server 2019 instance. The optimizer chooses this plan without any hints on my system configuration because the estimated cost of the plan is below the cost threshold for parallelism. Final Thoughts.

Servers 86

Servers Tuning Database Efficiency 86

SQL Performance

OCTOBER 29, 2019

DMV data is always available, so very often it's the easiest method to get a quick first look at query performance. Running this query on SQL Server 2014 and earlier requires removing reference to this DMV. In some scenarios, parameterizing five to 10 ad hoc queries can improve system performance as a whole. CROSS APPLY sys.

C++ 74

C++ Database Servers Tuning 74

Speed Curve

JUNE 10, 2015

The balance has shifted from just talking about how fast or reliable a particular system is to the overall experience a user has when they interact with and experience a product. Here’s Kadlec’s related talk from Velocity 2014 in New York: The path to performance. Successful performance begins far earlier than development.

Metrics 40

Metrics Design DevOps Website 40

Dynatrace

AUGUST 11, 2022

The FaaS model of cloud computing debuted in 2014 with startups like hook.io. Increased availability. Because FaaS is a cloud-native approach, it makes great use of multisite cloud architecture to improve availability and reliability. Consider the challenges of function as a service. These include the following: Reduced control.

Development 253

Development Serverless Best Practices Lambda 253

Brendan Gregg

SEPTEMBER 25, 2021

These strange questions came to the fore back in 2014 when Netflix was switching services from CentOS Linux to Ubuntu, and I helped debug several weird performance issues including one I'll describe here. CLI tools The Cassandra systems were EC2 virtual machine (Xen) instances. Microbenchmark os::javaTimeMillis() on both systems.

Speed 126

Speed Java AWS Virtualization 126

Dynatrace

SEPTEMBER 29, 2020

Since we moved to AWS in May 2014 we have had an availability of 99.95%! It was on August 25 th at 14:00 when Davis initially alerted on a disk write latency issues to Elastic File System (EFS) on one of our EC2 instances in AWS’s Sydney Data Center. Sydney, we have a disk write latency problem!

Infrastructure 181

Infrastructure Cloud Monitoring AWS 181

Dynatrace

MAY 31, 2021

Since Kubernetes emerged in 2014, it has become a popular solution for scaling, managing, and automating the deployments of containerized applications in distributed environments. Although Kubernetes simplifies application development while increasing resource utilization, it is a complex system that presents its own challenges.

Virtualization 133

Virtualization DevOps Infrastructure Cloud 133

Sutter's Mill

MARCH 11, 2024

We must make our software infrastructure more secure against the rise in cyberattacks (such as on power grids, hospitals, and banks), and safer against accidental failures with the increased use of software in life-critical systems (such as autonomous vehicles and autonomous weapons). The C++ community has not done so. best effort” analysis).

C++ 139

C++ Programming Code Java 139

Abhishek Tiwari

NOVEMBER 3, 2018

Recently I was asked about content management systems (CMS) of the future - more specifically how they are evolving in the era of microservices, APIs, and serverless computing. They often get blindsided by vendor’s pitch and end-up making decision based on some fancy demos (see my post from 2014 on Adobe AEM ).

Systems 63

Systems Cache Website Network 63

Alex Russell

APRIL 29, 2021

This post mines publicly available data on the pace of compatibility fixes and feature additions to assess the claim. After 20 years of neck-in-neck competition, often starting from common code lineages, there just isn't that much left to wring out of the system. Count of APIs available from JavaScript by Web Confluence.

Media 145

Media Games Education Engineering 145

Brendan Gregg

NOVEMBER 29, 2017

At Netflix, we've been using these technologies as they've been made available for instance types in the AWS EC2 cloud. In this configuration, the AMI and boot is paravirt (PV), the kernel is making hypercalls instead of privileged instructions, and the system is using paravirt network and storage drivers.

Virtualization 111

Virtualization AWS Hardware Storage 111

All Things Distributed

OCTOBER 25, 2018

The new AWS Africa (Cape Town) Region will have three Availability Zones and provide lower latency to end users across Sub-Saharan Africa. We already have 55 Availability Zones across 19 infrastructure regions that customers can use today. With the AWS based system, Exatype can do this at a fraction of the cost.

AWS 167

AWS Cloud Education Programming 167

The Netflix TechBlog

FEBRUARY 20, 2019

by Jason Koch , with Martin Spier , Brendan Gregg , Ed Hunter Improving the tools available to our engineers to help them diagnose, triage, and work through software performance challenges in the cloud is a key goal for the cloud performance engineering team at Netflix. a relatively old Angular 1.x x to React w/ Semantic UI.

Performance 211

Performance Latency Open Source Metrics 211

Scalegrid

JULY 17, 2020

Often times an external system is providing data as JSON, so it might be a temporary store before data is ingested into other parts of the system. 2014) added support for JSONB data type. For example, find all books that are available in braille. Managing High Availability in PostgreSQL. JSON database in 9.2

Storage 321

Storage Database Efficiency Processing 321

Dynatrace

FEBRUARY 25, 2021

Since its introduction by AWS in 2014, AWS Lambda has revolutionized the compute space and boosted the entire serverless movement. To handle N parallel requests, N Lambda instances need to be available, and AWS will spin up up to 1000 such instances automatically to handle 1000 parallel requests. Dynatrace news.

Lambda 221

Lambda AWS Monitoring Serverless 221

Brendan Gregg

SEPTEMBER 25, 2021

These strange questions came to the fore back in 2014 when Netflix was switching services from CentOS Linux to Ubuntu, and I helped debug several weird performance issues including one I'll describe here. CLI tools The Cassandra systems were EC2 virtual machine (Xen) instances. Microbenchmark os::javaTimeMillis() on both systems.

Speed 40

Speed Java AWS Virtualization 40

Brendan Gregg

SEPTEMBER 25, 2021

These strange questions came to the fore back in 2014 when Netflix was switching services from CentOS Linux to Ubuntu, and I helped debug several weird performance issues including one I'll describe here. CLI tools The Cassandra systems were EC2 virtual machine (Xen) instances. Microbenchmark os::javaTimeMillis() on both systems.

Speed 52

Speed Java AWS Virtualization 52

Alex Podelko

NOVEMBER 21, 2018

In 2014, I was preparing a presentation about load testing tools and criteria for their selection. It is great that the book is not limited by the core JMeter functionality and discusses available components and integrations when appropriate. The more complex the system, the more exotic multi-user performance problems can occur.

DevOps 113

DevOps Testing Testing Tools Performance Testing 113

All Things Distributed

APRIL 4, 2017

The new AWS EU (Stockholm) Region will have three Availability Zones and will be ready for customers to use in 2018. As a result, we have opened 42 Availability Zones across 16 AWS Regions worldwide. In 2014 and 2015 respectively, AWS opened offices in Stockholm and Espoo, Finland.

AWS 116

AWS Airlines Latency Games 116

All Things Distributed

DECEMBER 12, 2018

Today, I'm happy to announce that the AWS Europe (Stockholm) Region, our 20th Region globally, is now generally available for use by customers. With this launch, AWS now provides 60 Availability Zones, with another 12 zones and four Regions expected to come online by 2020 in Bahrain, Cape Town, Hong Kong, and Milan.

AWS 124

AWS Cloud Games Serverless 124

The Morning Paper

JANUARY 23, 2020

Today we’ll be digging into the analysis of an incident that took place at Etsy on December 4th, 2014. The shop had been closed so no data was available. First look for any correlation to the last change made to the system. This is part 2 of our look at Allspaw’s 2015 master thesis (here’s part 1 ).

Internet 53

Internet Internet Cache Engineering 53

All Things Distributed

JULY 20, 2015

A cluster is just a pool of compute resources available to a customer’s applications. The agent is written in Go, has a minimal footprint, and is available on GitHub under an Apache license. This architecture affords Amazon ECS high availability, low latency, and high throughput because the data store is never pessimistically locked.

Latency 166

Latency Architecture AWS Open Source 166