tail latency

Gil Tene of Azul Systems explains common pitfalls in measuring latency and response time, including why averages mislead and the problem of coordinated omission. He introduces open source tools like HdrHistogram for better measurement and reporting.

A short visualization from HotNets researchers of routing and latency over SpaceX's proposed Starlink constellation, based on its FCC filings. It shows orbital shells, link topology and how inter-satellite routes could beat terrestrial fiber on long distances.

John Carmack's keynote at Oculus Connect 2014 about the Gear VR and development stories from building it. He discusses mobile GPU optimization, latency, and display ideas like programmable interlacing.

Rick Hudson's GopherCon 2015 talk on Go 1.5's new low-latency concurrent garbage collector. It covers the motivation, performance numbers, technical challenges and why the approach fits Go.

A Google I/O 2013 talk by Android audio engineers on why reliable, low-latency audio is hard on the platform. It covers problem areas across the stack, platform work underway and techniques apps can use.

John Carmack speaks at UMKC's 2017 commencement event about the difficulties of VR, especially end-to-end latency across hardware and software. He covers why latency and buffering matter and how nausea is addressed.

Mark Handley of UCL explains, with simulation visualizations, how early Starlink satellites without inter-satellite links could still give low-latency wide-area communication by hopping via ground relays. He shows it can beat fibre on many long paths.

Euclideon's 2011 technology preview of Unlimited Detail, a rendering approach claiming unlimited geometric detail in real time via point-cloud style data. The company promised more to come while keeping most work secret.

Admiral Grace Hopper holds up a length of wire to show how far light travels in a nanosecond, about 11.8 inches. She explains how she made the idea concrete to people who could not picture a billionth of a second.

A two-minute 2008 clip from Sun's Fishworks team in which Brendan Gregg uses DTrace to show disk latency spiking while he shouts at a rack of drives. Bryan Cantrill narrates the demonstration.

Mark Handley of UCL looks at how SpaceX's November 2018 revisions to the Starlink constellation affect network behavior. The video visualizes the orbits and routing and estimates latencies, building on his earlier video and research paper.

A Microsoft Research demo from TechFest 2012 of a lab touch system that lets viewers feel the effect of different input latencies, down to about 1 ms. It shows how much more responsive dragging feels compared with typical 100 ms touch delay.

Optiver's Carl Cook explains how to build high-frequency trading systems in C++, where the critical path runs rarely but must be extremely fast. He covers techniques for low and predictable latency, including avoiding main memory and cache misses.

Real Engineering explains why SpaceX is building Starlink, including the economics of satellite internet. It covers the latency advantage of light in vacuum over fiber for long-distance links.

Redis Labs and Tim Davis present RedisGraph, which uses GraphBLAS to treat graph queries in Cypher as sparse linear algebra for lower latency.

A very long, detailed look at the A4/V2 rocket turbopump, the steam-turbine-driven twin propellant pump system. It covers the peroxide and permanganate catalysis that drives the turbine. This is part 1 of a series on V2 components.

Part 2 of a Q&A workshop with security researcher Jacob Appelbaum from April 2012, covering mobile phone security, cellphone forensics, state repression and surveillance. Topics include booting Tails from USB and surveillance self-defense resources.

A short comparison of the retail release of Metallica's Death Magnetic against the Guitar Hero 3 version of the same song. It switches between the two so you can hear and see the heavy clipping in the retail master.

Nassim Taleb argues that evidence on terrorism risk presented in media is statistically naive, using extreme value theory without math to separate fat-tailed risks like Ebola and terrorism from thin-tailed ones.

Full recording of Elon Musk's January 2015 presentation and Q&A at the SpaceX Seattle office announcement. He lays out plans for a satellite constellation to provide low-latency global internet, tied to funding Mars ambitions. Pauses between questions have been edited out.

A 2006 IBM commercial depicting a future supermarket where shoppers walk out and RFID automatically charges them for their items.

A short demo of typing on the Modos Paper Monitor, an open e-ink monitor, using 4-level grayscale to show low latency.

A one-minute clip from Who Framed Roger Rabbit in which a swinging pendant lamp is bumped and keeps flailing, showing animators' attention to detail.

A Swedish fiber provider's short experiment in which volunteers wear Oculus Rift headsets that add lag to their view of real life. They try to do everyday tasks like making pancakes.

Nicholas Ormrod's CppCon 2016 talk on how Facebook optimizes std::string with its open-source replacement fbstring. It covers storage layout tradeoffs, small string optimization and memory allocation behavior.

Jake Birkett of Grey Alien Games explains at GDC 2016 how an indie developer survived eleven years without a hit by building a long tail of games and redefining success.

Giant Bomb interviews John Carmack at E3 2012 about his virtual reality head-mounted display prototype. He demos it and talks about latency, depth and focus problems in VR.

Wendover Productions explains why American towns, retail and buildings increasingly look alike, covering zoning pressures, corporate standardization and cheap construction types like 5-over-1 buildings.

Grace Hopper appears on Letterman and uses a foot-long piece of wire to show how far light travels in a nanosecond.

Scott Galloway's DLD15 talk assessing the competitive positions of Amazon, Apple, Facebook and Google. He argues about where each is strong and weak, including Amazon's need for physical retail.

A lecture by Commodore Grace Hopper on the future of computing, posted by MIT Lincoln Laboratory. It includes her well-known nanosecond wire demonstration and humorous comparisons of device latencies.

Tom Scott visits IEX, a stock exchange that slows down trading with a long coil of fiber. The coil removes the speed advantage of high-frequency traders.

The original 2012 screencast introducing Meteor, a JavaScript framework for building web apps quickly. It demonstrates realtime updates, latency compensation and low boilerplate.

Munro Live's team breaks down the thermal management system of the Ford Mustang Mach-E and compares it with the Tesla Model Y. They look at parts count, coolant loops and heating approaches.

Ilya Grigorik of Google's Make the Web Fast team explains how to build a mobile site that appears on screen within one second. He covers the network and rendering constraints on mobile.

Gor Nishanov's CppCon 2018 talk on using tiny C++ coroutines to hide memory latency, interleaving useful work while prefetches bring cache lines in from main memory.

Brian Karis's SIGGRAPH 2021 talk explains how Unreal Engine 5's Nanite virtualized geometry works. It covers mesh import, the data structures, streaming, decompression, culling, software and hardware rasterization, and shading.

Zach Tellman's Clojure West talk covers the basics of queueing theory and its practical use in systems, including backpressure, how queues break down under load, and pitfalls of naive core.async use.

Dropbox engineer Kevin Modzelewski gives a Stanford colloquium on Dropbox's history and the technical problems it faced scaling, including detailed database schema decisions.

FortNine explains the economics of the dealership model, arguing that dealers absorb inventory risk and help manufacturers' finances. It looks at whether cars and motorcycles could be sold more cheaply without them.