Brendan Gregg

JULY 4, 2021

AWS Graviton2); for memory with the arrival of DDR5 and High Bandwidth Memory (HBM) on-processor; for storage including new uses for 3D Xpoint as a 3D NAND accelerator; for networking with the rise of QUIC and eXpress Data Path (XDP); and so on. Ford, et al., “TCP

Performance 121

Performance Latency Hardware Storage 121

Brendan Gregg

FEBRUARY 28, 2023

My personal opinion is that I don't see a widespread need for more capacity given horizontal scaling and servers that can already exceed 1 Tbyte of DRAM; bandwidth is also helpful, but I'd be concerned about the increased latency for adding a hop to more memory. Ford, et al., “TCP

Performance 103

Performance Latency Cache Virtualization 103

All Things Distributed

SEPTEMBER 25, 2017

This Region will consist of three Availability Zones at launch, and it will provide even lower latency to users across the Middle East. Plus another AWS GovCloud (US) Region in the United States is coming online by the end of 2018. This news marks the 22nd AWS Region we have announced globally.

Cloud 122

Cloud Education Energy Government 122

All Things Distributed

DECEMBER 12, 2018

They can run applications in Sweden, serve end users across the Nordics with lower latency, and leverage advanced technologies such as containers, serverless computing, and more. We launched Edge Network locations in Denmark, Finland, Norway, and Sweden. Our AWS Europe (Stockholm) Region is open for business now.

AWS 120

AWS Cloud Games Serverless 120

Abhishek Tiwari

DECEMBER 27, 2017

There are several emerging data trends that will define the future of ETL in 2018. In 2018, we anticipate that ETL will either lose relevance or the ETL process will disintegrate and be consumed by new data architectures. With the arrival of new cloud-native tools and platform, ETL is becoming obsolete. More details on this approach.

Big Data 49

Big Data Artificial Intelligence Storage Hardware 49

Alex Russell

JANUARY 30, 2024

It's time once again to update our priors regarding the global device and network situation. seconds on the target device and network profile, consuming 120KiB of critical path resources to become interactive, only 8KiB of which is script. What's changed since last year? and 75KiB of JavaScript. These are generous targets.

Performance 98

Performance Network Mobile Speed 98

Alex Russell

MARCH 6, 2021

Thanks to progress in networks and browsers (but not devices), a more generous global budget cap has emerged for sites constructed the "modern" way: ~100KiB of HTML/CSS/fonts and ~300-350KiB of JS (compressed) is the new rule-of-thumb limit for at least the next year or two. Modern network performance and availability.

Performance 127

Performance Network Metrics Cache 127

CSS - Tricks

APRIL 1, 2021

Google’s industry benchmarks from 2018 also provide a striking breakdown of how each second of loading affects bounce rates. Source: Google /SOASTA Research, 2018. Using a network request inspector, I’m going to see if there’s anything we can remove via the Network panel in DevTools.

Cache 92

Cache Social Media Media Network 92

Smashing Magazine

OCTOBER 31, 2023

The truth is that the two tools were fairly distinct until PSI was updated in 2018 to use Lighthouse reporting. Chrome DevTools includes a separate “Performance” tab where the testing environment’s CPU and network connection can be artificially throttled to mirror a specific testing condition, such as slow internet speeds.

Speed 100

Speed Google Website Metrics 100

ACM Sigarch

MARCH 18, 2019

How many buffers are needed to track pending requests as a function of needed bandwidth and expected latency? Can one both minimize latency and maximize throughput for unscheduled work? These models are useful for insight regarding the basic computer system performance metrics of latency and throughput (bandwidth). Little’s Law.

Systems 60

Systems Latency Performance Analytics 60

Brendan Gregg

JULY 4, 2021

AWS Graviton2); for memory with the arrival of DDR5 and High Bandwidth Memory (HBM) on-processor; for storage including new uses for 3D Xpoint as a 3D NAND accelerator; for networking with the rise of QUIC and eXpress Data Path (XDP); and so on. Ford, et al., “TCP

Performance 52

Performance Latency Hardware Storage 52

Brendan Gregg

OCTOBER 7, 2018

eBPF was created by Alexei Starovoitov while at PLUMgrid (he's now at Facebook) as a generic in-kernel virtual machine, with software defined networks as the primary use case. Here's key differences as of August 2018: Type DTrace bpftrace. It's shaping up to be a DTrace version 2.0: eBPF does more. variable this->name $name.

C++ 40

C++ Virtualization Programming Latency 40

KeyCDN

OCTOBER 24, 2018

Today we’re excited to announce that we’ve launched yet another POP location to help further supercharge our network’s content delivery speeds. Although both countries are relatively close to one another, they are separated by a distance of approximately 500km, which adds up in terms of latency. What’s Next?

Internet 48

Internet Internet Speed Network 48

ACM Sigarch

MARCH 25, 2019

How many buffers are needed to track pending requests as a function of needed bandwidth and expected latency? Can one both minimize latency and maximize throughput for unscheduled work? The M/M/1 queue will show us a required trade-off among (a) allowing unscheduled task arrivals, (b) minimizing latency, and (c) maximizing throughput.

Systems 53

Systems Latency Performance C++ 53

Brendan Gregg

AUGUST 18, 2019

For example, iostat(1), or a monitoring agent, may tell you your average disk latency, but not the distribution of this latency. For smaller environments, it can be of more use helping eliminate latency outliers. Block I/O latency as a histogram. 6 * 7 * Copyright 2018 Netflix, Inc. 1 /* 2 * biolatency.bt

Latency 57

Latency C++ Cache Programming 57

Speed Curve

FEBRUARY 5, 2024

Charlie Vazac introduced server timing in a Performance Calendar post circa 2018. The use of server-timing headers by content delivery networks closes a big gap. Latency – How much time does it take to deliver a packet from A to B. In practice, it looks something like this: server-timing: processing_time; dur=123.4;

Servers 60

Servers Cache Retail Benchmarking 60

Alex Russell

OCTOBER 22, 2017

We constrain ourselves to a real-world baseline device + network configuration to measure progress. Budgets are scaled to a benchmark network & device. JavaScript is the single most expensive part of any page in ways that are a function of both network capacity and device speed. The median user is on a slow network.

Performance 74

Performance Network Benchmarking Mobile 74

Brendan Gregg

FEBRUARY 28, 2023

My personal opinion is that I don't see a widespread need for more capacity given horizontal scaling and servers that can already exceed 1 Tbyte of DRAM; bandwidth is also helpful, but I'd be concerned about the increased latency for adding a hop to more memory. Ford, et al., “TCP

Performance 52

Performance Latency Cache Virtualization 52

Alex Russell

SEPTEMBER 11, 2018

I can count on one hand the number of teams I’ve worked with who have goals that allow them to block launches for latency regressions, including Google products. — Alex Russell (@slightlylate) August 27, 2018. Few teams I’ve encountered have actionable metrics associated with the real experiences of their users.

Development 73

Development Government Mobile Network 73

Alex Russell

APRIL 29, 2021

Real-time network protocols for enabling videoconferencing, desktop sharing, and game streaming applications. After years of standards discussion and the first delivered to other platforms in 2018, iOS 14.5 April 2018 , but not usable until several releases later). Delayed five years. Gamepad API. Pointer Lock. Media Recorder.

Media 145

Media Games Education Engineering 145

KeyCDN

SEPTEMBER 12, 2019

An easy way to compress images is with our image processing service that happens to also be fully integrated into our existing network. Image CDN Using a content delivery network like KeyCDN, or what we also call an image CDN , can be one of the easiest and fastest ways to speed up the delivery of your images.

Social Media 106

Social Media Media Google Website 106

Smashing Magazine

SEPTEMBER 6, 2021

Next, we’ll look at how to set up servers and clients (that’s the hard part unless you’re using a content delivery network (CDN)). Using just a few (but still more than one), however, could nicely balance congestion growth with better performance, especially on high-speed networks. Servers and Networks. Network Configuration.

Network 108

Network Servers Cache Traffic 108

Abhishek Tiwari

DECEMBER 27, 2017

In 2018, a widespread adaptation of Kubernetes for big data processing is anitcipated. optimised container networking and security. direct access to raw block storage [18] without the abstraction of a filesystem for workloads that require consistent I/O performance and low latency. overall performance comparable to bare metal.

Big Data 63

Big Data Storage Benchmarking Hardware 63

Adrian Cockcroft

NOVEMBER 18, 2019

OPN304 Learnings from migrating a service from JDK 8 to JDK 11 AWS Lambda improved latency by migrating to JDK 11 with Amazon Corretto. OPN402 Firecracker open-source innovation Since Firecracker’s release at re:Invent 2018, several open-source teams have built on it, while AWS has continued investing in Firecracker’s speed.

Open Source 40

Open Source AWS Lambda Serverless 40

Adrian Cockcroft

NOVEMBER 18, 2019

OPN304 Learnings from migrating a service from JDK 8 to JDK 11 AWS Lambda improved latency by migrating to JDK 11 with Amazon Corretto. OPN402 Firecracker open-source innovation Since Firecracker’s release at re:Invent 2018, several open-source teams have built on it, while AWS has continued investing in Firecracker’s speed.

Open Source 40

Open Source AWS Lambda Serverless 40

Dean Hume

APRIL 24, 2018

Each HTTP request that is required for the page needs to travel over the network and in turn this consumes energy on both the server and client. Without effective caching on the client, the server will see an increase in workload, more CPU usage and ultimately increased latency for the end user. It’s a win-win all round!

Energy 40

Energy Cache Traffic Website 40

Dean Hume

APRIL 24, 2018

Each HTTP request that is required for the page needs to travel over the network and in turn this consumes energy on both the server and client. Without effective caching on the client, the server will see an increase in workload, more CPU usage and ultimately increased latency for the end user. It’s a win-win all round!

Energy 40

Energy Cache Traffic Website 40

Dean Hume

APRIL 24, 2018

Each HTTP request that is required for the page needs to travel over the network and in turn this consumes energy on both the server and client. Without effective caching on the client, the server will see an increase in workload, more CPU usage and ultimately increased latency for the end user. It’s a win-win all round!

Energy 40

Energy Cache Traffic Website 40

IO River

NOVEMBER 2, 2023

Delta Air Lines experienced a severe system outage in 2017, resulting in flight cancellations and delays across their network. Proactive monitoring aids in detecting performance bottlenecks, latency difficulties, and other anomalies that may influence availability.

Availability 52

Availability Social Media Traffic Games 52

IO River

NOVEMBER 2, 2023

Delta Air Lines experienced a severe system outage in 2017, resulting in flight cancellations and delays across their network. Proactive monitoring aids in detecting performance bottlenecks, latency difficulties, and other anomalies that may influence availability.

Availability 52

Availability Social Media Traffic Games 52

All Things Distributed

JUNE 20, 2017

The new AWS Asia Pacific (Hong Kong) Region will have three Availability Zones and be ready for customers for use in 2018. As well as AWS Regions, we also have 21 AWS Edge Network Locations in Asia Pacific. This enables customers to serve content to their end users with low latency, giving them the best application experience.

AWS 127

AWS Logistics Cloud Social Media 127

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 well as AWS Regions, we also have 24 AWS Edge Network Locations in Europe. This enables customers to serve content to their end users with low latency, giving them the best application experience.

AWS 110

AWS Airlines Latency Games 110

Smashing Magazine

JANUARY 6, 2020

Globally in 2018–2019, according to the IDC, 87% of all shipped mobile phones are Android devices. If you don’t have a device at hand, emulate mobile experience on desktop by testing on a throttled 3G network (e.g. Estimated Input Latency tells us if we are hitting that threshold, and ideally, it should be below 50ms.

Performance 112

Performance Cache Network Servers 112

Testsigma

FEBRUARY 12, 2021

The growing demand for IoT-testing is the government’s gradual acceptance of smart cities’ concept, which is why businesses are keen to incorporate IoT into their networks. Gartner stated that Artificial Intelligence computing resources are to expand by 5 times between 2018 and 2023. Autonomous Test Automation.

Artificial Intelligence 92

Artificial Intelligence Software Software IoT 92

Smashing Magazine

JANUARY 7, 2019

If you don’t have a device at hand, emulate mobile experience on desktop by testing on a throttled network (e.g. To make the performance impact more visible, you could even introduce 2G Tuesdays or set up a throttled 3G network in your office for faster testing. 150ms RTT, 1.5 Mbps down, 0.7 Mbps up) with a throttled CPU (5× slowdown).

Performance 105

Performance Cache Metrics Network 105

Smashing Magazine

JANUARY 11, 2021

Networking, HTTP/2, HTTP/3 OCSP stapling, EV/DV certificates, packaging, IPv6, QUIC, HTTP/3. If you don’t have a device at hand, emulate mobile experience on desktop by testing on a throttled 3G network (e.g. Estimated Input Latency tells us if we are hitting that threshold, and ideally, it should be below 50ms. 300ms RTT, 1.6

Performance 145

Performance Cache Media Metrics 145