Daniel Lemire is a software performance expert. He ranks among the top 2% of scientists globally (Stanford/Elsevier 2025) and is one of GitHub's top 1000 most followed developers.
Visit lemire.me → RSS feedOn Hacker News May 2014 – Sep 2026
Software Engineering 47Systems & Performance 40Math & Algorithms 14Hardware & Electronics 14Society, Policy & Culture 12Tech Industry & Business 8
People still tell me that CPUs are boring. That nothing much happens anymore. Let us look at AMD Ryzen 7 processors from 2022 to 2024: the 5800X3D (Zen 3), the 7800X3D (Zen 4) and the 9800X3D (Zen 5). They are comparable 8-core chips with 3D V-Cache. I have written about them before, in How stagnant … Continue reading…
In software, you frequently need to check whether some objects is in a set. For example, you might have a list of forbidden Web addresses. As someone enters a new Web address, you may want to check whether it is part of your black list. Or maybe you have a large list of already used … Continue reading Xor Filters…
Some of the latest Intel processors support the AVX-512 family of vector instructions. These instructions operate on blocks of 512 bits (or 64 bytes). The benefit of such wide instructions is that even without increasing the processor clock speed, systems can still process a lot more data. Most code today operators…
When a program is mostly just accessing memory randomly, a standard cost model is to count the number of distinct random accesses. The general idea is that memory access is much slower than most other computational tasks. Furthermore, the cost model can be extended to count "nearby" memory accesses as free. That is…
The number of researchers and peer-review publications is growing exponentially. It has been estimated that the number of researchers in the world doubles every 16 years and the number of research outputs is increasing even faster. If you accept that published research papers are an accurate measure of our scientific…
We sometimes have to look for a value in a sorted array. The simplest algorithm consists in just going through the values We sometimes have to look for a value in a sorted array. The simplest algorithm consists in just going through the values one by one, until we encounter the value, or exhaust the … Continue reading…
I am convinced that much of the gap between the best college students and the worst is explained by study habits. Frankly, most students study poorly. To make matters worse, most teachers are incapable of teaching good study habits. Learning is proportional with effort Sitting in a classroom listening to a professor…
As a kid, my parents would open the television set, and we would get to watch whatever the state television decided we would watch. It was a push model. Some experts pick the content you need and they deliver it to you. You have little say in the matter. There was one newspaper in my … Continue reading Stop spending…
Programmers often need to 'filter out' data. Suppose that you are given a database of users where only a small percentage are 'paying customers' (say 5% or less). You can write an SQL query to check whether a given user is indeed a paying customer, but it might require a round trip to your database … Continue reading…
One of the most common "data type" in programming is the text string. When programmers think of a string, they imagine that they are dealing with a list or an array of characters. It is often a "good enough" approximation, but reality is more complex. The characters must be encoded into bits in some way. Most ……
Last year (2019), we released the simjson library. It is a C++ library available under a liberal license (Apache) that can parse JSON documents very fast. How fast? We reach and exceed 3 gigabytes per second in many instances. It can also parse millions of small JSON documents per second. The new version is much ……
In my previous blog post, I compared the performance of my new ARM-based MacBook Pro with my 2017 Intel-based MacBook Pro. I used a number parsing benchmark. In some cases, the ARM-based MacBook Pro was nearly twice as fast as the older Intel-based MacBook Pro. I think that the Apple M1 processor is a breakthrough ……
I envy economists. Unlike computer scientists, they seem to be able to publish best-seller books with innovative research. One such book is Piketty's Capital. The book is reminiscent of Marx's capital in its scope. If you haven't heard about the book yet, it has a simple message: the yield on capital is higher than…
Modern processors have instructions to process several bytes at once. Effectively all processors have the capability of processing 16 bytes at once. These instructions are called SIMD, for single instruction, multiple data. It was once an open question whether these instructions could be useful to accelerate common…
When programming, we often need to write 'generic' functions where the exact data type is not important. For example, you might want to write a simple function that sums up numbers. Go lacked this notion until recently, but it was recently added (as of version 1.18). So I took it out for a spin. In … Continue reading…
Let us say that I ask you to find the number I am thinking about between -1000 and 1000, by repeatedly guessing a number. With each guess, I tell you whether your guess is correct, smaller or larger than my number. A binary search algorithm tries to find a value in an interval by repeating … Continue reading Fast…
If I multiply two 64-bit integers (having values in [0, 264)), the product requires 128 bits. Intel and AMD processors (x64) can compute the full (128-bit) product of two 64-bit integers using a single instruction (mul). ARM processors, such as those found in your mobile phone, require two instructions to achieve the…
In software, it is common to represent time as a time-stamp string. It is usually specified by a time format string. Some standards use the format %Y%m%d%H%M%S meaning that we print the year, the month, the day, the hours, the minutes and the seconds. The current time as I write this blog post would be … Continue…
In software programming, the product between two integers is often computed to a fixed number of bits with overflow. Consider 8-bit integers. If you multiply 127 by 127, you get back the number 1 as an 8-bit unsigned integer, with an overflow. The actual full product is 16129. To represent 16129, you typically use 16…
“We see something that works, and then we understand it.” (Thomas Dullien) It is a deeper insight than it seems. Young people spend years in school learning the reverse: understanding happens before progress. That is the linear theory of innovation. So Isaac Newton comes up with his three laws of mechanics, and we get…
One of the established and most popular programming languages is the C programming language. It is relatively easy to learn, and highly practical. Maybe surprisingly, the C programming language keeps evolving, slowly and carefully. If you have GCC 13 or LLVM (Clang) 16, you already have a compiler with some of the…
Computers store data as streams of bits. Binary files like image, audio or video files are allowed to contain just about any sequence of bits. However, we also often use text formats; for example, web pages and emails are required to be text. So how do we send images by email? How do we embed … Continue reading…
Sometimes programmers want to prune out characters from a string of characters. For example, maybe you want to remove all line-ending characters from a piece of text. Let me consider the problem where I want to remove all spaces (' ') and linefeed characters ('\n' and '\r'). How would you do it efficiently? size_t…
Most commodity processors support single-precision IEEE 754 floating-point numbers. Though they are ubiquitous, they are often misunderstood. One of my readers left a comment suggesting that picking an integer in [0,232) at random and dividing it by 232, was equivalent to picking a number at random in [0,1). I am not…
Not all instructions on modern processors cost the same. Additions and subtractions are cheaper than multiplications which are themselves cheaper than divisions. For this reason, compilers frequently replace division instructions by multiplications. Roughly speaking, it works in this manner. Suppose that you want to…
Science is the belief in the ignorance of experts said Richard Feynman. Feynman had a Nobel prize in physics. He was a remarquable educator: his lecture notes are still popular. He foresaw nanotechnology and quantum computing. He is credited with identifying the cause of the Space Shuttle Challenger disaster. There is…
Almost all our computers are made of several processing cores. Thus it can be efficient to "parallelize" expensive processing in a multicore manner. That is, instead of using a single core to do all of the work, you divide the work among multiple cores. A standard way to approach this problem is to create threads. ……
When benchmarking software, we often start by measuring the time elapsed. If you are benchmarking data bandwidth or latency, it is the right measure. However, if you are benchmarking computational tasks where you avoid disk and network accesses and where you only access a few pages of memory, then the time elapsed is…
Developers often believe that software performance follows a Pareto distribution: 80% of the running time is spent in 20% of the code. Using this model, you can write most of your code without any care for performance and focus on the narrow pieces of code that are performance sensitive. Engineers like Casey Muratori…
In the C programming language, we allocate memory dynamically (on the heap) using the malloc function. You pass malloc a size parameter corresponding to the number of bytes you need. The function returns either a pointer to the allocated memory or the NULL pointer if the memory could not be allocated. Or so you may ……