Brendan Gregg

SEPTEMBER 25, 2021

A Cassandra database cluster had switched to Ubuntu and noticed write latency increased by over 30%. As a Xen guest, this profile was gathered using perf(1) and the kernel's software cpu-clock soft interrupts, not the hardware NMI. A quick check of basic performance statistics showed over 30% higher CPU consumption.

Speed 126

Speed Java AWS Virtualization 126

Alex Russell

MARCH 6, 2021

TL;DR: A lot has changed since 2017 when we last estimated a global baseline resource per-page resource budget of 130-170KiB. To update our global baseline from 2017, we want to update our priors on a few dimensions: The evolved device landscape. Hardware Past As Performance Prologue. The Moto G4 , for example. Hard Reset.

Performance 124

Performance Network Cache Metrics 124

Brendan Gregg

SEPTEMBER 25, 2021

A Cassandra database cluster had switched to Ubuntu and noticed write latency increased by over 30%. As a Xen guest, this profile was gathered using perf(1) and the kernel's software cpu-clock soft interrupts, not the hardware NMI. A quick check of basic performance statistics showed over 30% higher CPU consumption.

Speed 40

Speed Java AWS Virtualization 40

Alex Russell

JANUARY 30, 2024

HTML, CSS, images, and fonts can all be parsed and run at near wire speeds on low-end hardware, but JavaScript is at least three times more expensive, byte-for-byte. If you or your company are able to generate a credible worldwide latency estimate in the higher percentiles for next year's update, please get in touch.

Performance 90

Performance Network Mobile Speed 90

Alex Russell

APRIL 29, 2021

As an engineer on a browser team, I'm privy to the blow-by-blow of various performance projects, benchmark fire drills, and the ways performance marketing (deeply) impacts engineering priorities. With each team, benchmarks lost are understood as bugs. WebGL 2 launched for other platforms on Chrome and Firefox in 2017.

Media 145

Media Games Education Engineering 145

John McCalpin

JANUARY 22, 2018

Introduction: In December 2017, my colleague Damon McDougall (now at AMD) asked for help in porting the fused multiply-add example code from a Colfax report ( [link] ) to the Xeon Phi x200 (Knights Landing) processors here at TACC. A peculiar throughput limitation on Intel’s Xeon Phi x200 (Knights Landing).

Latency 40

Latency Hardware Code Testing 40

Alex Russell

OCTOBER 22, 2017

— Alex Russell (@slightlylate) October 4, 2017. Budgets are scaled to a benchmark network & device. Deciding what benchmark to use for a performance budget is crucial. It simulates a link with a 400ms RTT and 400-600Kbps of throughput (plus latency variability and simulated packet loss). Global Ground-Truth.

Performance 68

Performance Benchmarking Network Mobile 68