Techreport blog

Engineering Insights & Guides

The best problem-solving writing in software engineering — databases, performance, security, frontend and distributed systems, in one feed.

11,670 articles
Engineering — Q2 2024 Internet disruption summary

Q2 2024 Internet disruption summary

Government directed shutdowns and cable cuts were both significant sources of Internet outages in Q2 2024. This post explores these disruptions, as well as others caused by power outages, maintenance,

DBDavid BelsonDavid Belson·July 16, 2024Engineering
SRE & Ops — AI Lab: The secrets to keeping machine learning engineers moving fast

AI Lab: The secrets to keeping machine learning engineers moving fast

The key to developer velocity across AI lies in minimizing time to first batch (TTFB) for machine learning (ML) engineers. AI Lab is a pre-production framework used internally at Meta. It allows us to continuously A/B test common ML workflows – enabling proactive improvements and automatically preventing regressions on TTFB. AI Lab prevents TTFB regressions […]

CWChris Wiltz·July 16, 2024SRE & Ops
Security — Application Security report: 2024 update

Application Security report: 2024 update

Cloudflare’s updated 2024 view on Internet cyber security trends spanning global traffic insights, bot traffic insights, API traffic insights, and client-side risks

MTMichael Tremante, SabinaMichael Tremante, Sabina·July 11, 2024Security
AI & ML — Taming the tail utilization of ads inference at Meta scale

Taming the tail utilization of ads inference at Meta scale

Tail utilization is a significant system issue and a major factor in overload-related failures and low compute utilization. The tail utilization optimizations at Meta have had a profound impact on model serving capacity footprint and reliability. Failure rates, which are mostly timeout errors, were reduced by two-thirds; the compute footprint delivered 35% more work for […]

CWChris Wiltz·July 10, 2024AI & ML
SRE & Ops — Meta’s approach to machine learning prediction robustness

Meta’s approach to machine learning prediction robustness

Meta’s advertising business leverages large-scale machine learning (ML) recommendation models that power millions of ads recommendations per second across Meta’s family of apps. Maintaining reliability of these ML systems helps ensure the highest level of service and uninterrupted benefit delivery to our users and advertisers. To minimize disruptions and ensure our ML systems are intrinsically

CWChris Wiltz·July 10, 2024SRE & Ops
Design & UX — When Friction Is A Good Thing: Designing Sustainable E-Commerce Experiences — Smashing Magazine

When Friction Is A Good Thing: Designing Sustainable E-Commerce Experiences — Smashing Magazine

Today, UX design contributes to the problem of excessive consumption through persuasive e-commerce practices and designing for companies with linear and exploitative business models. Anna Rátkai explores practical steps we can take to reduce material consumption and the massive environmental impact that comes along with it by designing mindful shopping experiences.

ARAnna RatkaiAnna Ratkai·July 10, 2024Design & UX
Security — DDoS threat report for 2024 Q2

DDoS threat report for 2024 Q2

Welcome to the 18th edition of the Cloudflare DDoS Threat Report. Released quarterly, these reports provide an in-depth analysis of the DDoS threat landscape as observed across the Cloudflare network. This edition focuses on the second quarter of 2024

OJOmer, JorgeOmer, Jorge·July 9, 2024Security
Security — RADIUS/UDP vulnerable to improved MD5 collision attack

RADIUS/UDP vulnerable to improved MD5 collision attack

The RADIUS protocol is commonly used to control administrative access to networking gear. Despite its importance, RADIUS hasn’t changed much in decades. We discuss an attack on RADIUS as a case study for why it’s important for legacy protocols to keep up with advancements in cryptography

GMGoldbe, Miro HallerGoldbe, Miro Haller·July 9, 2024Security
Frontend — "If" CSS Gets Inline Conditionals

"If" CSS Gets Inline Conditionals

A few sirens went off a couple of weeks ago when the CSS Working Group (CSSWG) resolved to add an if() conditional to the CSS Values Module Level 5

GGGeoff GrahamGeoff Graham·July 9, 2024Frontend
Networking — Cloudflare 1.1.1.1 incident on June 27, 2024

Cloudflare 1.1.1.1 incident on June 27, 2024

On June 27, 2024, a small number of users globally may have noticed that 1.1.1.1 was unreachable or degraded. The root cause was a mix of BGP (Border Gateway Protocol) hijacking and a route leak

BMBryton, MingweiBryton, Mingwei·July 4, 2024Networking