A Crash Course in Modern Hardware by Cliff Click
Why Hacker News recommends it
Recommended as an accessible refresher on how modern hardware and caches affect software performance.
What it is
Cliff Click walks through a small performance example on a modern out-of-order CPU. He covers pipelining, ILP, CISC vs RISC, multicore, and why cache misses and the memory subsystem dominate performance.
Recommended by 3 HN commenters across 3 different months, 2017 to 2025.
What Hacker News said
Recommends zellyn · 2017-03 · on “Cold, Hard Cache – Insomniac Games’ Cache Simulator [slides]”
Coincidentally, I just watched this last night: it's a really accessible refresher if you're a bit outdated on how all these caches work and interact: [this video]
Recommends tjalfi · 2021-06 · on “Intel to disable TSX by default on more CPUs with new microcode”
I'm also a fan of Cliff and other Rice alumni; here are a few of his videos that I like. [0] The Sea of Nodes and the HotSpot JIT [1] A JVM Does That? [2] A Crash Course in Modern Hardware [3] Bits of Advice for the VM Writer [0] [LINK] [1] [LINK] [2] [this video] [3] [LINK]
Recommends NovaX · 2025-03 · on “Microbenchmarks: HashMap, ConcurrentHashMap, and Guava Caches”
This was a good attempt but flawed. 1. You should use JMH to handle warmup, jit, timings, averaging of runs, etc. You might enjoy the following talks on the subject, though I am sure there are other gems out there. - Performance Anxiety: [LINK] - Benchmarking for Good: [LINK] - (The Art of) (Java) Benchmarking: [LINK] - A Crash Course in Modern Hardware: [this video] - How NOT to Write a Microbenchmark [LINK] - Anatomy of a flawed microbenchmark [LINK] 2. An uncontended lock is basically free, as you noticed. 3. Guava does implement the Map interface via asMap(). It also have a concurrencyLevel to adjust its performance. It is based on Java 5's CHM. 4. Caffeine is a Guava Cache rewrite based on J
Cites dr_zoidberg · 2017-10 · on “Why Python is Slow: Looking Under the Hood (2014)”
While the name of the talk implies it's a generally-speaking-talk, Click goes into a lot of detail of how the JVM works in his "A Crash Course in Modern Hardware" presentation[0]. I saw it a while ago, but I think he does mention those things. Of course that's the one I saw and remember, but you could probably find another presentation in which he goes into much better detail about the JVM and its inner workings. [0] [this video]
Cites belter · 2021-09 · on “IBM mainframe beats other platforms for cost and performance”
Finding a bit strange the overall tone of this thread. Running a mainframe comes with the challenges of finding specialists, relying on a single vendor, having to buy at least two in two different data centers for disaster recovery,etc...But purely from performance point of view, as in throughput, and also quality of service, the case is clear for the mainframe. Spend some time reading the Z15 technical guide, and marvel at what is today the state of the art in computing. We are talking about sustained, full on 5.2 GHz, all the time. Massive IO bandwith, ECC memory everywhere, massive caches, native crypto co-processors. All this, while your x86, most likely, spends most of its time waiting
Cites xk3 · 2024-06 · on “The Engine of the Future”
I understood the main points pretty well and the reasoning behind most of them. I don't work in the gaming industry. I draw upon these resources for my context: - [LINK] - Mike Acton "Data-Oriented Design and C++" [LINK] - A Crash Course in Modern Hardware by Cliff Click [this video] - GCAP 2017: Decima Visibility in HZD [LINK] - GPU-Based Run-Time Procedural Placement in HZD [LINK] - [LINK] I say this only to make the point that optimization is not completely inaccessible to the average person. It is useful to understand what the hardware is doing--up to a certain point; good abstractions are also necessary.