/Performance

Where Exactly Does Python 3.11 Get Its ~25% Speedup?

- Beshr Kayali tl;dr: "Python 3.11 was released a few days ago and as usual, it comes with more than a few new features that make it ever more interesting, from exception groups and fine-grained error locations and tracebacks to TOML parsing support in the standard library and of course the much awaited speedup as part of the faster CPython project. CPython 3.11 is 25% faster than CPython 3.10 on average according to benchmarks with pyperformance."

featured in #367


Prioritizing App Stability - Mobile Performance @ Lyft

- Wen Zhao tl;dr: We focused our investment in mobile performance into the 3 metrics with the highest opportunity for improvement: (1) Time to interact: continuing reducing app startup time. (2) Stability: reducing the number of crashes any given user experiences. (3) Rendering performance: maintaining a high, buttery smooth frame rate. 

featured in #358


Why Your Website Should Be Under 14B In Size 

tl;dr: "Having a smaller website makes it load faster — that's not surprising. What is surprising is that a 14kB page can load much faster than a 15kB page — maybe 612ms faster — while the difference between a 15kB and a ;16kB page is trivial. This is because of the TCP slow start algorithm. This article will cover what that is, how it works, and why you should care. But first we'll quickly go over some of the basics."

featured in #346


Python 3.11 Is Up To 10 - 60% Faster Than Python 3.10

tl;dr: "CPython 3.11 is on average 25% faster than CPython 3.10 when measured with the pyperformance benchmark suite, and compiled with GCC on Ubuntu Linux. Depending on your workload, the speedup could be up to 10-60% faster. This project focuses on two major areas in Python: faster startup and faster runtime."

featured in #332


The Two Generals Problem

- Seth Archer Brown tl;dr: "Imagine there’s a city in a valley. On either side of the valley, there’s an army commanded by a general. On one side stands General Alice and her army. On the other, General Bob and his army. Alice and Bob want to capture the city, but neither has an army large enough to do so alone. Alice and Bob must attack at the same time, but Alice and Bob can only communicate by sending messengers through the valley, who have a chance of being captured by the city’s army. How do Alice and Bob coordinate their attack?"

featured in #319


How To Speed Up ClickHouse Queries Using Materialized Columns

- Karl-Aksel Puulmann tl;dr: ClickHouse supports speeding up queries using materialized columns to create new columns on the fly from existing data. In this post, I’ll walk through a query optimization example that's well-suited to this rarely-used feature.

featured in #307


The Fastest GIF Does Not Exist

- Ben Phelps tl;dr: "If you're here because you want to fix your GIF and want the quick answer, the solution is: set your frame delay to 20ms instead of 10ms. If you want to learn a bit more about GIFs, exactly why this edge case happens, and some thoughts on how to improve things, keep reading!"

featured in #293


Fixing Performance Regressions Before They Happen

tl;dr: “This post describes how the Netflix TVUI team implemented a robust strategy to quickly and easily detect performance anomalies before they are released — and often before they are even committed to the codebase.”

featured in #288


Memory Leaks: The Forgotten Side Of Web Performance

- Nolan Lawson tl;dr: "If you speed up your website by 2 seconds, everyone agrees that that’s a good thing with a visible user impact. If you reduce your website’s memory leak by 2 MB, can we still agree it was worth it? Maybe not." Nolan discusses the unique characteristics of memory leaks: (1) Low-impact until critical. (2) Hard to diagnose. (3) Trivial to fix once diagnosed.

featured in #280


The Speed Of Time

- Brendan Gregg tl;dr: Brendan describes a weird performance issue he debugged: "A Cassandra database cluster had switched to Ubuntu and noticed write latency increased by over 30%. A quick check of basic performance statistics showed over 30% higher CPU consumption. What on Earth is Ubuntu doing that results in 30% higher CPU time!?" 

featured in #255