Optimization on Producthunt daily

1brc

Mon, 01 Sep 2025 15:29:43 +0800

gunnarmorling/1brc

1️⃣🐝🏎️ The One Billion Row Challenge

Status Feb 4: The final leaderboards have been published. Congrats to all the winners, and a big thank you to everyone participating in this challenge as well as to everyone helping to organize it!

Status Feb 3: All entries have been evaluated and I am in the process of finalizing the leaderboards.

Status Feb 1: The challenge has been closed for new submissions. No new pull requests for adding submissions are accepted at this time. Pending PRs will be evaluated over the next few days.

Status Jan 31: The challenge will close today at midnight UTC.

Status Jan 12: As there has been such a large number of entries to this challenge so far (100+), and this is becoming hard to manage, please only create new submissions if you expect them to run in 10 seconds or less on the evaluation machine.

Status Jan 1: This challenge is open for submissions!

Sponsorship

A big thank you to my employer Decodable for funding the evaluation environment and supporting this challenge!

The One Billion Row Challenge (1BRC) is a fun exploration of how far modern Java can be pushed for aggregating one billion rows from a text file. Grab all your (virtual) threads, reach out to SIMD, optimize your GC, or pull any other trick, and create the fastest implementation for solving this task!

The text file contains temperature values for a range of weather stations. Each row is one measurement in the format <string: station name>;<double: measurement>, with the measurement value having exactly one fractional digit. The following shows ten rows as an example:

Hamburg;12.0
Bulawayo;8.9
Palembang;38.8
St. John's;15.2
Cracow;12.6
Bridgetown;26.9
Istanbul;6.2
Roseau;34.4
Conakry;31.2
Istanbul;23.0

The task is to write a Java program which reads the file, calculates the min, mean, and max temperature value per weather station, and emits the results on stdout like this (i.e. sorted alphabetically by station name, and the result values per station in the format <min>/<mean>/<max>, rounded to one fractional digit):

`1`	`{Abha=-23.0/18.0/59.2, Abidjan=-16.2/26.0/67.3, Abéché=-10.0/29.4/69.0, Accra=-10.1/26.4/66.4, Addis Ababa=-23.7/16.0/67.0, Adelaide=-27.8/17.3/58.5, ...}`

Submit your implementation by Jan 31 2024 and become part of the leaderboard!

Results

These are the results from running all entries into the challenge on eight cores of a Hetzner AX161 dedicated server (32 core AMD EPYC™ 7502P (Zen2), 128 GB RAM).

#	Result (m:s.ms)	Implementation	JDK	Submitter	Notes	Certificates
1	00:01.535	link	21.0.2-graal	Thomas Wuerthinger, Quan Anh Mai, Alfonso² Peterssen	GraalVM native binary, uses Unsafe	Certificate
2	00:01.587	link	21.0.2-graal	Artsiom Korzun	GraalVM native binary, uses Unsafe	Certificate
3	00:01.608	link	21.0.2-graal	Jaromir Hamala	GraalVM native binary, uses Unsafe	Certificate
	00:01.880	link	21.0.1-open	Serkan ÖZAL	uses Unsafe	Certificate
	00:01.921	link	21.0.2-graal	Van Phu DO	GraalVM native binary, uses Unsafe	Certificate
	00:02.018	link	21.0.2-graal	Stephen Von Worley	GraalVM native binary, uses Unsafe	Certificate
	00:02.157	link	21.0.2-graal	Roy van Rijn	GraalVM native binary, uses Unsafe	Certificate
	00:02.319	link	21.0.2-graal	Yavuz Tas	GraalVM native binary, uses Unsafe	Certificate
	00:02.332	link	21.0.2-graal	Marko Topolnik	GraalVM native binary, uses Unsafe	Certificate
	00:02.367	link	21.0.1-open	Quan Anh Mai	uses Unsafe	Certificate
	00:02.507	link	21.0.1-open	gonix	uses Unsafe	Certificate
	00:02.557	link	21.0.1-open	yourwass	uses Unsafe	Certificate
	00:02.820	link	22.ea.32-open	Li Lin	uses Unsafe	Certificate
	00:02.995	link	21.0.2-graal	tivrfoa	GraalVM native binary, uses Unsafe	Certificate
	00:02.997	link	21.0.1-open	gonix		Certificate
	00:03.095	link	21.0.2-graal	Jamal Mulla	GraalVM native binary, uses Unsafe	Certificate
	00:03.210	link	21.0.1-open	Quan Anh Mai		Certificate
	00:03.298	link	21.0.1-graal	Subrahmanyam	uses Unsafe	Certificate
	00:03.431	link	21.0.1-graal	Roman Musin	GraalVM native binary, uses Unsafe	Certificate
	00:03.469	link	21.0.2-graal	Elliot Barlas	GraalVM native binary, uses Unsafe	Certificate
	00:03.698	link	21.0.1-graal	Jason Nochlin		Certificate
	00:03.785	link	21.0.2-graal	zerninv	GraalVM native binary, uses Unsafe	Certificate
	00:03.820	link	21.0.2-graal	John Ziamos	GraalVM native binary, uses Unsafe	Certificate
	00:03.902	link	21.0.1-open	Juan Parera		Certificate
	00:03.966	link	21.0.1-open	Jin Cong Ho	uses Unsafe	Certificate
	00:03.991	link	21.0.1-graal	Vaidhy Mayilrangam	uses Unsafe	Certificate
	00:04.066	link	21.0.1-open	JesseVanRooy	uses Unsafe	Certificate
	00:04.101	link	21.0.2-graal	Jaime Polidura	GraalVM native binary, uses Unsafe	Certificate
	00:04.209	link	21.0.1-open	Giovanni Cuccu		Certificate
	00:04.474	link	21.0.1-open	Roman Stoffel		Certificate
	00:04.676	link	21.0.2-tem	Peter Levart		Certificate
	00:04.684	link	21.0.1-open	Florin Blanaru	uses Unsafe	Certificate
	00:04.701	link	21.0.1-open	Dr Ian Preston		Certificate
	00:04.741	link	21.0.1-open	Cliff Click	uses Unsafe	Certificate
	00:04.800	link	21.0.1-open	Parker Timmins		Certificate
	00:04.884	link	21.0.1-open	Aleksey Shipilëv		Certificate
	00:04.920	link	21.0.1-graal	Subrahmanyam		Certificate
	00:05.077	link	21.0.2-graal	Jonathan Wright	GraalVM native binary, uses Unsafe	Certificate
	00:05.142	link	21.0.1-open	Arjen Wisse		Certificate
	00:05.167	link	21.0.2-open	Yevhenii Melnyk		Certificate
	00:05.235	link	21.0.1-open	unbounded		Certificate
	00:05.336	link	java	Sumit Chaudhary	uses Unsafe	Certificate
	00:05.354	link	21.0.2-graal	Arman Sharif	GraalVM native binary, uses Unsafe	Certificate
	00:05.478	link	21.0.1-open	Olivier Bourgain	uses Unsafe	Certificate
	00:05.559	link	21.0.1-graal	Panagiotis Drakatos	GraalVM native binary	Certificate
	00:05.887	link	21.0.1-graal	Charlie Evans	uses Unsafe	Certificate
	00:05.979	link	21.0.1-graal	Sam Pullara		Certificate
	00:06.166	link	21.0.1-open	Jamie Stansfield		Certificate
	00:06.257	link	21.0.1-graal	Stefan Sprenger	uses Unsafe	Certificate
	00:06.392	link	21.0.2-graal	Diego Parra		Certificate
	00:06.576	link	21.0.1-open	Andrew Sun	uses Unsafe	Certificate
	00:06.635	link	21.0.1-graal	Laake Scates-Gervasi	GraalVM native binary, uses Unsafe	Certificate
	00:06.654	link	21.0.1-graal	Jaroslav Bachorik		Certificate
	00:06.715	link	21.0.1-open	Algirdas Raščius		Certificate
	00:06.884	link	21.0.1-graal	rcasteltrione		Certificate
	00:06.982	link	21.0.1-open	Chris Bellew		Certificate
	00:07.563	link	21.0.1-graal	3j5a		Certificate
	00:07.680	link	21.0.1-graal	Xylitol	uses Unsafe	Certificate
	00:07.712	link	21.0.1-graal	Anita SV		Certificate
	00:07.730	link	21.0.1-open	Johannes Schüth		Certificate
	00:07.894	link	21.0.2-tem	Antonio Muñoz		Certificate
	00:07.925	link	21.0.1-graal	Ricardo Pieper		Certificate
	00:07.948	link	java	Smoofie	uses Unsafe	Certificate
	00:08.157	link	21.0.1-open	JurenIvan		Certificate
	00:08.167	link	21.0.1-tem	Dimitar Dimitrov		Certificate
	00:08.214	link	21.0.1-open	deemkeen		Certificate
	00:08.255	link	21.0.1-open	Mathias Bjerke		Certificate
	00:08.398	link	21.0.1-open	Parth Mudgal	uses Unsafe	Certificate
	00:08.489	link	21.0.1-graal	Bang NGUYEN		Certificate
	00:08.517	link	21.0.1-graal	ags	uses Unsafe	Certificate
	00:08.557	link	21.0.1-graal	Adrià Cabeza		Certificate
	00:08.622	link	21.0.1-graal	Keshavram Kuduwa	uses Unsafe	Certificate
	00:08.892	link	21.0.1-open	Roman Romanchuk		Certificate
	00:08.896	link	21.0.1-open	Andrzej Nestoruk		Certificate
	00:09.020	link	21.0.1-open	yemreinci		Certificate
	00:09.071	link	21.0.1-open	Gabriel Reid		Certificate
	00:09.352	link	21.0.1-graal	Filip Hrisafov		Certificate
	00:09.725	link	21.0.2-graal	Martin	GraalVM native binary	Certificate
	00:09.867	link	21.0.1-graal	Ricardo Pieper		Certificate
	00:09.945	link	21.0.1-open	Anthony Goubard		Certificate
	00:10.092	link	21.0.1-graal	Pratham		Certificate
	00:10.127	link	21.0.1-open	Parth Mudgal	uses Unsafe	Certificate
	00:11.577	link	21.0.1-open	Eve		Certificate
	00:10.473	link	21.0.1-open	Anton Rybochkin		Certificate
	00:11.119	link	21.0.1-open	lawrey		Certificate
	00:11.156	link	java	Yann Moisan		Certificate
	00:11.167	link	21.0.1-open	Nick Palmer		Certificate
	00:11.352	link	21.0.1-open	karthikeyan97	uses Unsafe	Certificate
	00:11.363	link	21.0.2-tem	Guruprasad Sridharan		Certificate
	00:11.405	link	21.0.1-graal	Rafael Merino García		Certificate
	00:11.406	link	21.0.1-graal	gabrielfoo		Certificate
	00:11.433	link	21.0.1-graal	Jatin Gala		Certificate
	00:11.505	link	21.0.1-open	Dmitry Bufistov	uses Unsafe	Certificate
	00:11.744	link	21.0.2-tem	Sebastian Lövdahl		Certificate
	00:11.805	link	21.0.1-graal	Cool_Mineman		Certificate
	00:11.934	link	21.0.1-open	arjenvaneerde		Certificate
	00:12.220	link	21.0.1-open	Richard Startin		Certificate
	00:12.495	link	21.0.1-graal	Samuel Yvon	GraalVM native binary	Certificate
	00:12.568	link	21.0.1-graal	Vlad		Certificate
	00:12.800	link	java	Yonatan Graber		Certificate
	00:13.013	link	21.0.1-graal	Thanh Duong		Certificate
	00:13.071	link	21.0.1-open	Dr Ian Preston		Certificate
	00:13.729	link	java	Cedric Boes		Certificate
	00:13.817	link	21.0.1-open	Carlo		Certificate
	00:14.502	link	21.0.1-graal	eriklumme		Certificate
	00:14.772	link	21.0.1-open	Kevin McMurtrie		Certificate
	00:14.867	link	21.0.1-open	Michael Berry		Certificate
	00:14.900	link	java	Judekeyser		Certificate
	00:15.006	link	java	Paweł Adamski		Certificate
	00:15.662	link	21.0.1-open	Serghei Motpan		Certificate
	00:16.063	link	21.0.1-open	Marek Kohn		Certificate
	00:16.457	link	21.0.1-open	Aleksei		Certificate
	00:16.953	link	21.0.1-open	Gaurav Anantrao Deshmukh		Certificate
	00:17.046	link	21.0.1-open	Dimitris Karampinas		Certificate
	00:17.086	link	java	Breejesh Rathod		Certificate
	00:17.490	link	21.0.1-open	Gergely Kiss		Certificate
	00:17.255	link	21.0.1-open	tkosachev		Certificate
	00:17.520	link	21.0.1-open	Farid		Certificate
	00:17.717	link	21.0.1-open	Oleh Marchenko		Certificate
	00:17.815	link	21.0.1-open	Hallvard Trætteberg		Certificate
	00:17.932	link	21.0.1-open	Bartłomiej Pietrzyk		Certificate
	00:18.251	link	21.0.1-graal	Markus Ebner		Certificate
	00:18.448	link	21.0.1-open	Moysés Borges Furtado		Certificate
	00:18.771	link	21.0.1-graal	David Kopec		Certificate
	00:18.902	link	21.0.1-graal	Maxime		Certificate
	00:19.357	link	21.0.1-graalce	Roman Schweitzer		Certificate
	00:20.691	link	21.0.1-graal	Kidlike	GraalVM native binary	Certificate
	00:21.989	link	21.0.1-open	couragelee		Certificate
	00:22.188	link	21.0.1-open	Jairo Graterón		Certificate
	00:22.334	link	21.0.1-open	Alberto Venturini		Certificate
	00:22.457	link	21.0.1-open	Ramzi Ben Yahya		Certificate
	00:22.471	link	21.0.1-open	Shivam Agarwal		Certificate
	00:24.986	link	21.0.1-open	kumarsaurav123		Certificate
	00:25.064	link	21.0.2-open	Sudhir Tumati		Certificate
	00:26.500	link	21.0.1-open	Bruno Félix		Certificate
	00:28.381	link	21.0.1-open	Hampus		Certificate
	00:29.741	link	21.0.1-open	Matteo Vaccari		Certificate
	00:32.018	link	21.0.1-open	Aurelian Tutuianu		Certificate
	00:34.388	link	21.0.1-tem	Tobi		Certificate
	00:35.875	link	21.0.1-open	MahmoudFawzyKhalil		Certificate
	00:36.180	link	21.0.1-open	Horia Chiorean		Certificate
	00:36.424	link	java	Manish Garg		Certificate
	00:38.340	link	21.0.1-open	AbstractKamen		Certificate
	00:41.982	link	21.0.1-open	Chris Riccomini		Certificate
	00:42.893	link	21.0.1-open	javamak		Certificate
	00:46.597	link	21.0.1-open	Maeda-san		Certificate
	00:58.811	link	21.0.1-open	Ujjwal Bharti		Certificate
	01:05.094	link	21.0.1-open	Mudit Saxena		Certificate
	01:05.979	link	21.0.1-graal	Hieu Dao Quang		Certificate
	01:06.790	link	21.0.1-open	Karl Heinz Marbaise		Certificate
	01:06.944	link	21.0.1-open	santanu		Certificate
	01:07.014	link	21.0.1-open	pedestrianlove		Certificate
	01:07.101	link	21.0.1-open	Jeevjyot Singh Chhabda		Certificate
	01:08.811	link	21.0.1-open	Aleš Justin		Certificate
	01:08.908	link	21.0.1-open	itaske		Certificate
	01:09.595	link	21.0.1-tem	Antonio Goncalves		Certificate
	01:09.882	link	21.0.1-open	Prabhu R		Certificate
	01:14.815	link	21.0.1-open	twohardthings		Certificate
	01:25.801	link	21.0.1-open	ivanklaric		Certificate
	01:33.594	link	21.0.1-open	Gaurav Mathur		Certificate
	01:53.208	link	java	Mahadev K		Certificate
	01:56.607	link	21.0.1-open	Abhilash		Certificate
	03:43.521	link	21.0.1-open	김예환 Ye-Hwan Kim (Sam)		Certificate
	03:59.760	link	21.0.1-open	Samson		Certificate
	—
	04:49.679	link (Baseline)	21.0.1-open	Gunnar Morling

Note that I am not super-scientific in the way I’m running the contenders (see Evaluating Results for the details). This is not a high-fidelity micro-benchmark and there can be variations of up to +-3% between runs. So don’t be too hung up on the exact ordering of your entry compared to others in close proximity. The primary purpose of this challenge is to learn something new, have fun along the way, and inspire others to do the same. The leaderboard is only means to an end for achieving this goal. If you observe drastically different results though, please open an issue.

See Entering the Challenge for instructions how to enter the challenge with your own implementation. The Show & Tell features a wide range of 1BRC entries built using other languages, databases, and tools.

Bonus Results

This section lists results from running the fastest N entries with different configurations. As entries have been optimized towards the specific conditions of the original challenge description and set-up (such as size of the key set), challenge entries may perform very differently across different configurations. These bonus results are provided here for informational purposes only. For the 1BRC challenge, only the results in the previous section are of importance.

32 Cores / 64 Threads

For officially evaluating entries into the challenge, each contender is run on eight cores of the evaluation machine (AMD EPYC™ 7502P). Here are the results from running the top 50 entries (as of commit e1fb378a, Feb 2) on all 32 cores / 64 threads (i.e. SMT is enabled) of the machine:

#	Result (m:s.ms)	Implementation	JDK	Submitter	Notes
1	00:00.323	link	21.0.2-graal	Jaromir Hamala	GraalVM native binary, uses Unsafe
2	00:00.326	link	21.0.2-graal	Thomas Wuerthinger, Quan Anh Mai, Alfonso² Peterssen	GraalVM native binary, uses Unsafe
3	00:00.349	link	21.0.2-graal	Artsiom Korzun	GraalVM native binary, uses Unsafe
	00:00.351	link	21.0.2-graal	Van Phu DO	GraalVM native binary, uses Unsafe
	00:00.389	link	21.0.2-graal	Stephen Von Worley	GraalVM native binary, uses Unsafe
	00:00.408	link	21.0.2-graal	Yavuz Tas	GraalVM native binary, uses Unsafe
	00:00.415	link	21.0.2-graal	Roy van Rijn	GraalVM native binary, uses Unsafe
	00:00.499	link	21.0.2-graal	Marko Topolnik	GraalVM native binary, uses Unsafe
	00:00.602	link	21.0.1-graal	Roman Musin	GraalVM native binary, uses Unsafe
	00:00.623	link	21.0.1-open	gonix	uses Unsafe
	00:00.710	link	21.0.2-graal	Jamal Mulla	GraalVM native binary, uses Unsafe
	00:00.727	link	21.0.2-graal	tivrfoa	GraalVM native binary, uses Unsafe
	00:00.774	link	21.0.1-open	Serkan ÖZAL	uses Unsafe
	00:00.788	link	21.0.2-graal	Elliot Barlas	GraalVM native binary, uses Unsafe
	00:00.832	link	21.0.2-graal	zerninv	GraalVM native binary, uses Unsafe
	00:00.840	link	21.0.1-open	gonix
	00:00.857	link	21.0.2-graal	Jaime Polidura	GraalVM native binary, uses Unsafe
	00:00.880	link	21.0.2-graal	John Ziamos	GraalVM native binary, uses Unsafe
	00:00.939	link	21.0.1-open	Aleksey Shipilëv
	00:01.026	link	21.0.1-open	JesseVanRooy	uses Unsafe
	00:01.118	link	21.0.2-graal	Jonathan Wright	GraalVM native binary
	00:01.140	link	21.0.2-graal	Arman Sharif	GraalVM native binary, uses Unsafe
	00:01.143	link	21.0.1-open	Cliff Click	uses Unsafe
	00:01.169	link	21.0.2-open	Yevhenii Melnyk
	00:01.188	link	21.0.1-graal	Subrahmanyam	uses Unsafe
	00:01.193	link	21.0.1-open	Florin Blanaru	uses Unsafe
	00:01.234	link	21.0.1-open	Olivier Bourgain	uses Unsafe
	00:01.242	link	21.0.1-open	Quan Anh Mai	uses Unsafe
	00:01.252	link	21.0.1-open	Jin Cong Ho	uses Unsafe
	00:01.267	link	22.ea.32-open	Li Lin	uses Unsafe
	00:01.363	link	21.0.2-tem	Peter Levart
	00:01.380	link	21.0.1-graal	Jason Nochlin
	00:01.391	link	21.0.1-open	Quan Anh Mai
	00:01.439	link	21.0.1-open	Arjen Wisse
	00:01.446	link	21.0.1-open	Dr Ian Preston
	00:01.504	link	21.0.1-open	Jamie Stansfield
	00:01.514	link	21.0.1-graal	Subrahmanyam
	00:01.516	link	21.0.1-graal	Vaidhy Mayilrangam	uses Unsafe
	00:01.586	link	21.0.1-open	yourwass	uses Unsafe
	00:01.647	link	21.0.2-graal	Diego Parra
	00:01.694	link	21.0.1-open	Parker Timmins
	00:01.694	link	21.0.1-graal	Charlie Evans	uses Unsafe
	00:01.702	link	21.0.1-graal	Sam Pullara
	00:01.733	link	java	Sumit Chaudhary	uses Unsafe
	00:01.742	link	21.0.1-open	unbounded
	00:02.241	link	21.0.1-graal	Stefan Sprenger	uses Unsafe
	00:02.294	link	21.0.1-open	Giovanni Cuccu
	00:02.990	link	21.0.1-graal	Panagiotis Drakatos	GraalVM native binary
	00:03.205	link	21.0.1-open	Juan Parera
	00:10.929	link	21.0.1-open	Roman Stoffel

10K Key Set

The 1BRC challenge data set contains 413 distinct weather stations, whereas the rules allow for 10,000 different station names to occur. Here are the results from running the top 40 entries (as of commit e1fb378a, Feb 2) against 1,000,000,000 measurement values across 10K stations (created via ./create_measurements3.sh 1000000000), using eight cores on the evaluation machine:

#	Result (m:s.ms)	Implementation	JDK	Submitter	Notes
1	00:02.957	link	21.0.2-graal	Artsiom Korzun	GraalVM native binary, uses Unsafe
2	00:03.058	link	21.0.2-graal	Marko Topolnik	GraalVM native binary, uses Unsafe
3	00:03.186	link	21.0.2-graal	Stephen Von Worley	GraalVM native binary, uses Unsafe
	00:03.998	link	21.0.2-graal	Roy van Rijn	GraalVM native binary, uses Unsafe
	00:04.042	link	21.0.2-graal	Jaromir Hamala	GraalVM native binary, uses Unsafe
	00:04.289	link	21.0.1-open	gonix	uses Unsafe
	00:04.522	link	21.0.2-graal	tivrfoa	GraalVM native binary, uses Unsafe
	00:04.653	link	21.0.2-graal	Jamal Mulla	GraalVM native binary, uses Unsafe
	00:04.733	link	21.0.1-open	gonix
	00:04.836	link	21.0.1-graal	Subrahmanyam	uses Unsafe
	00:04.870	link	21.0.2-graal	Thomas Wuerthinger, Quan Anh Mai, Alfonso² Peterssen	GraalVM native binary, uses Unsafe
	00:05.240	link	21.0.2-graal	zerninv	GraalVM native binary, uses Unsafe
	00:05.394	link	21.0.2-graal	Yavuz Tas	GraalVM native binary, uses Unsafe
	00:05.906	link	21.0.2-graal	Elliot Barlas	GraalVM native binary, uses Unsafe
	00:06.086	link	21.0.2-graal	Van Phu DO	GraalVM native binary, uses Unsafe
	00:06.379	link	21.0.2-graal	John Ziamos	GraalVM native binary, uses Unsafe
	00:07.113	link	21.0.2-open	Yevhenii Melnyk
	00:07.542	link	21.0.2-graal	Jonathan Wright	GraalVM native binary
	00:07.889	link	21.0.1-open	Florin Blanaru	uses Unsafe
	00:07.970	link	21.0.1-open	Cliff Click	uses Unsafe
	00:08.857	link	21.0.1-open	Serkan ÖZAL
	00:09.333	link	21.0.1-open	yourwass	uses Unsafe
	00:09.722	link	21.0.1-open	Aleksey Shipilëv
	00:09.777	link	21.0.1-graal	Vaidhy Mayilrangam	uses Unsafe
	00:10.263	link	21.0.1-open	Quan Anh Mai	uses Unsafe
	00:11.154	link	21.0.1-open	Parker Timmins
	00:13.175	link	21.0.1-open	Quan Anh Mai
	00:13.245	link	21.0.1-open	Dr Ian Preston
	00:13.377	link	21.0.1-open	Giovanni Cuccu
	00:13.761	link	21.0.1-open	Juan Parera
	00:14.441	link	21.0.2-tem	Peter Levart
	00:15.548	link	21.0.1-open	Jin Cong Ho	uses Unsafe
	00:17.906	link	21.0.1-graal	Jason Nochlin
	00:18.770	link	22.ea.32-open	Li Lin	uses Unsafe
	00:19.106	link	21.0.1-open	Roman Stoffel
	00:20.151	link	21.0.1-graal	Roman Musin	GraalVM native binary, uses Unsafe; seg-faults occassionally
	00:22.953	link	21.0.2-graal	Jaime Polidura	GraalVM native binary, uses Unsafe
	—
	DNF	link	21.0.1-open	JesseVanRooy	Incorrect output
	DNF	link	21.0.1-graal	Subrahmanyam	Doesn’t complete in 60 sec
	DNF	link	21.0.1-open	Arjen Wisse	Incorrect output

Prerequisites

Java 21 must be installed on your system.

Running the Challenge

This repository contains two programs:

dev.morling.onebrc.CreateMeasurements (invoked via create_measurements.sh): Creates the file measurements.txt in the root directory of this project with a configurable number of random measurement values
dev.morling.onebrc.CalculateAverage (invoked via calculate_average_baseline.sh): Calculates the average values for the file measurements.txt

Execute the following steps to run the challenge:

Build the project using Apache Maven:
1

./mvnw clean verify
Create the measurements file with 1B rows (just once):
1

./create_measurements.sh 1000000000
This will take a few minutes. Attention: the generated file has a size of approx. 12 GB, so make sure to have enough diskspace.

If you’re running the challenge with a non-Java language, there’s a non-authoritative Python script to generate the measurements file at src/main/python/create_measurements.py. The authoritative method for generating the measurements is the Java program dev.morling.onebrc.CreateMeasurements.
Calculate the average measurement values:
1

./calculate_average_baseline.sh
The provided naive example implementation uses the Java streams API for processing the file and completes the task in ~2 min on environment used for result evaluation. It serves as the base line for comparing your own implementation.
Optimize the heck out of it:

Adjust the CalculateAverage program to speed it up, in any way you see fit (just sticking to a few rules described below). Options include parallelizing the computation, using the (incubating) Vector API, memory-mapping different sections of the file concurrently, using AppCDS, GraalVM, CRaC, etc. for speeding up the application start-up, choosing and tuning the garbage collector, and much more.

Flamegraph/Profiling

A tip is that if you have jbang installed, you can get a flamegraph of your program by running async-profiler via ap-loader:

jbang --javaagent=ap-loader@jvm-profiling-tools/ap-loader=start,event=cpu,file=profile.html -m dev.morling.onebrc.CalculateAverage_yourname target/average-1.0.0-SNAPSHOT.jar

or directly on the .java file:

jbang --javaagent=ap-loader@jvm-profiling-tools/ap-loader=start,event=cpu,file=profile.html src/main/java/dev/morling/onebrc/CalculateAverage_yourname

When you run this, it will generate a flamegraph in profile.html. You can then open this in a browser and see where your program is spending its time.

Rules and limits

Any of these Java distributions may be used:
- Any builds provided by SDKMan
- Early access builds available on openjdk.net may be used (including EA builds for OpenJDK projects like Valhalla)
- Builds on builds.shipilev.net If you want to use a build not available via these channels, reach out to discuss whether it can be considered.
No external library dependencies may be used
Implementations must be provided as a single source file
The computation must happen at application runtime, i.e. you cannot process the measurements file at build time (for instance, when using GraalVM) and just bake the result into the binary
Input value ranges are as follows:
- Station name: non null UTF-8 string of min length 1 character and max length 100 bytes, containing neither ; nor \n characters. (i.e. this could be 100 one-byte characters, or 50 two-byte characters, etc.)
- Temperature value: non null double between -99.9 (inclusive) and 99.9 (inclusive), always with one fractional digit
There is a maximum of 10,000 unique station names
Line endings in the file are \n characters on all platforms
Implementations must not rely on specifics of a given data set, e.g. any valid station name as per the constraints above and any data distribution (number of measurements per station) must be supported
The rounding of output values must be done using the semantics of IEEE 754 rounding-direction “roundTowardPositive”

Entering the Challenge

To submit your own implementation to 1BRC, follow these steps:

Create a fork of the onebrc GitHub repository.
Run ./create_fork.sh <your_GH_user> to copy the baseline implementation to your personal files, or do this manually:
- Create a copy of CalculateAverage_baseline.java, named CalculateAverage_<your_GH_user>.java, e.g. CalculateAverage_doloreswilson.java.
- Create a copy of calculate_average_baseline.sh, named calculate_average_<your_GH_user>.sh, e.g. calculate_average_doloreswilson.sh.
- Adjust that script so that it references your implementation class name. If needed, provide any JVM arguments via the JAVA_OPTS variable in that script. Make sure that script does not write anything to standard output other than calculation results.
- (Optional) OpenJDK 21 is used by default. If a custom JDK build is required, create a copy of prepare_baseline.sh, named prepare_<your_GH_user>.sh, e.g. prepare_doloreswilson.sh. Include the SDKMAN command sdk use java [version] in the your prepare script.
- (Optional) If you’d like to use native binaries (GraalVM), add all the required build logic to your prepare_<your_GH_user>.sh script.
Make that implementation fast. Really fast.
Run the test suite by executing /test.sh <your_GH_user>; if any differences are reported, fix them before submitting your implementation.
Create a pull request against the upstream repository, clearly stating
- The name of your implementation class.
- The execution time of the program on your system and specs of the same (CPU, number of cores, RAM). This is for informative purposes only, the official runtime will be determined as described below.
I will run the program and determine its performance as described in the next section, and enter the result to the scoreboard.

Note: I reserve the right to not evaluate specific submissions if I feel doubtful about the implementation (I.e. I won’t run your Bitcoin miner ;).

If you’d like to discuss any potential ideas for implementing 1BRC with the community, you can use the GitHub Discussions of this repository. Please keep it friendly and civil.

The challenge runs until Jan 31 2024. Any submissions (i.e. pull requests) created after Jan 31 2024 23:59 UTC will not be considered.

Evaluating Results

Results are determined by running the program on a Hetzner AX161 dedicated server (32 core AMD EPYC™ 7502P (Zen2), 128 GB RAM).

Programs are run from a RAM disk (i.o. the IO overhead for loading the file from disk is not relevant), using 8 cores of the machine. Each contender must pass the 1BRC test suite (/test.sh). The hyperfine program is used for measuring execution times of the launch scripts of all entries, i.e. end-to-end times are measured. Each contender is run five times in a row. The slowest and the fastest runs are discarded. The mean value of the remaining three runs is the result for that contender and will be added to the results table above. The exact same measurements.txt file is used for evaluating all contenders. See the script evaluate.sh for the exact implementation of the evaluation steps.

Prize

If you enter this challenge, you may learn something new, get to inspire others, and take pride in seeing your name listed in the scoreboard above. Rumor has it that the winner may receive a unique 1️⃣🐝🏎️ t-shirt, too!

FAQ

Q: Can I use Kotlin or other JVM languages other than Java?
A: No, this challenge is focussed on Java only. Feel free to inofficially share implementations significantly outperforming any listed results, though.

Q: Can I use non-JVM languages and/or tools?
A: No, this challenge is focussed on Java only. Feel free to inofficially share interesting implementations and results though. For instance it would be interesting to see how DuckDB fares with this task.

Q: I’ve got an implementation—but it’s not in Java. Can I share it somewhere?
A: Whilst non-Java solutions cannot be formally submitted to the challenge, you are welcome to share them over in the Show and tell GitHub discussion area.

Q: Can I use JNI?
A: Submissions must be completely implemented in Java, i.e. you cannot write JNI glue code in C/C++. You could use AOT compilation of Java code via GraalVM though, either by AOT-compiling the entire application, or by creating a native library (see here.

Q: What is the encoding of the measurements.txt file?
A: The file is encoded with UTF-8.

Q: Can I make assumptions on the names of the weather stations showing up in the data set?
A: No, while only a fixed set of station names is used by the data set generator, any solution should work with arbitrary UTF-8 station names (for the sake of simplicity, names are guaranteed to contain no ; or \n characters).

Q: Can I copy code from other submissions?
A: Yes, you can. The primary focus of the challenge is about learning something new, rather than “winning”. When you do so, please give credit to the relevant source submissions. Please don’t re-submit other entries with no or only trivial improvements.

Q: Which operating system is used for evaluation?
A: Fedora 39.

Q: My solution runs in 2 sec on my machine. Am I the fastest 1BRC-er in the world?
A: Probably not :) 1BRC results are reported in wallclock time, thus results of different implementations are only comparable when obtained on the same machine. If for instance an implementation is faster on a 32 core workstation than on the 8 core evaluation instance, this doesn’t allow for any conclusions. When sharing 1BRC results, you should also always share the result of running the baseline implementation on the same hardware.

Q: Why 1️⃣🐝🏎️ ?
A: It’s the abbreviation of the project name: One Billion Row Challenge.

1BRC on the Web

A list of external resources such as blog posts and videos, discussing 1BRC and specific implementations:

The One Billion Row Challenge Shows That Java Can Process a One Billion Rows File in Two Seconds , by Olimpiu Pop (interview)
Cliff Click discussing his 1BRC solution on the Coffee Compiler Club (video)
1️⃣🐝🏎️🦆 (1BRC in SQL with DuckDB), by Robin Moffatt (blog post)
1 billion rows challenge in PostgreSQL and ClickHouse, by Francesco Tisiot (blog post)
The One Billion Row Challenge with Snowflake, by Sean Falconer (blog post)
One billion row challenge using base R, by David Schoch (blog post)
1 Billion Row Challenge with Apache Pinot, by Hubert Dulay (blog post)
One Billion Row Challenge In C, by Danny Van Kooten (blog post)
One Billion Row Challenge in Racket, by Bogdan Popa (blog post)
The One Billion Row Challenge - .NET Edition, by Frank A. Krueger (podcast)
One Billion Row Challenge, by Ragnar Groot Koerkamp (blog post)
ClickHouse and The One Billion Row Challenge, by Dale McDiarmid (blog post)
One Billion Row Challenge & Azure Data Explorer, by Niels Berglund (blog post)
One Billion Row Challenge - view from sidelines, by Leo Chashnikov (blog post)
1 billion row challenge in SQL and Oracle Database, by Gerald Venzl (blog post)
One Billion Row Challenge: Learned So Far, by Roman Stoffel (blog post)
One Billion Row Challenge in Racket, by Bogdan Popa (blog post)
The 1 Billion row challenge with Singlestore, by Anna Semjen (blog post)
1BRC in .NET among fastest on Linux: My Optimization Journey, by Victor Baybekov (blog post)
One Billion Rows – Gerald’s Challenge, by Connor McDonald (blog post)
Reading a file insanely fast in Java, by Romain Manni-Bucau (blog post)
#1BRC Timeline, by tivrfoa (blog post)
1BRC - What a Journey, by Marius Staicu (blog post)
One Billion Rows Challenge in Golang, by Shraddha Agrawal (blog post)
The Billion Row Challenge (1BRC) - Step-by-step from 71s to 1.7s by Marko Topolnik (blog post)
Entering The One Billion Row Challenge With GitHub Copilot by Antonio Goncalves (blog post)
DataFrame and The One Billion Row Challenge–How to use a Java DataFrame to save developer time, produce readable code, and not win any prizes by Vladimir Zakharov (blog post)

License

This code base is available under the Apache License, version 2.

Code of Conduct

Be excellent to each other! More than winning, the purpose of this challenge is to have fun and learn something new.

Resume-Matcher

Sat, 26 Jul 2025 15:30:22 +0800

srbhr/Resume-Matcher

Resume Matcher

𝙹𝚘𝚒𝚗 𝙳𝚒𝚜𝚌𝚘𝚛𝚍 ✦ 𝚆𝚎𝚋𝚜𝚒𝚝𝚎 ✦ 𝙷𝚘𝚠 𝚝𝚘 𝙸𝚗𝚜𝚝𝚊𝚕𝚕 ✦ 𝙲𝚘𝚗𝚝𝚛𝚒𝚋𝚞𝚝𝚘𝚛𝚜 ✦ 𝙳𝚘𝚗𝚊𝚝𝚎 ✦ 𝚃𝚠𝚒𝚝𝚝𝚎𝚛/𝚇 ✦ 𝙻𝚒𝚗𝚔𝚎𝚍𝙸𝚗

Stop getting auto-rejected by ATS bots. Resume Matcher is the AI-powered platform that reverse-engineers hiring algorithms to show you exactly how to tailor your resume. Get the keywords, formatting, and insights that actually get you past the first screen and into human hands.

Hoping to make this, VS Code for making resumes.

[!IMPORTANT]

This project is in active development. New features are being added continuously, and we welcome contributions from the community. There are some breaking changes on the main branch. If you have any suggestions or feature requests, please feel free to open an issue on GitHub or discuss it on our Discord server.

Getting started with Resume Matcher

Resume Matcher is designed to help you optimize your resume with the aim to highlight your skills and experience in a way that resonates with potential employers.

We’re working on actively improving the platform, towards building a VS Code for making resumes, and adding new features. The best way to stay updated is to join the discord discussion and be part of the active development community.

Join our Discord community 👇

Follow us on LinkedIn ✨

⭐ Star Resume Matcher to support the development and get updates on GitHub.

Key Features

Works locally: No need to upload your resume to a server. Everything runs on your machine with open source AI models by Ollama.
ATS Compatibility: Get a detailed analysis of your resume’s compatibility with ATS systems.
Instant Match Score: Upload resume & job description for a quick match score and key improvement areas.
Keyword Optimizer: Align your resume with job keywords and identify critical content gaps.
Guided Improvements: Get clear suggestions to make your resume stand out.

Roadmap

If you have any suggestions or feature requests, please feel free to open an issue on GitHub. And discuss it on our Discord server.

Visual keyword highlighting.
AI Canvas, which can help to craft impactful, metric-driven resume content.
Multi-job description optimization.

How to Install

Follow the instructions in the SETUP.md file to set up the project locally. The setup script will install all the necessary dependencies and configure your environment.

The project is built using:

FastAPI for the backend.
Next.js for the frontend.
Ollama for local AI model serving.
Tailwind CSS for styling.
SQLite for the database.

Technology	Info/Version
Python	3.12+
Next.js	15+
Ollama	0.6.7

Join Us and Contribute

We welcome contributions from everyone! Whether you’re a developer, designer, or just someone who wants to help out. All the contriburtors are listed in the about page on our website and on the GitHub Readme here.

Checkout the roadmap if you would like to work on the features that are planned for the future. If you have any suggestions or feature requests, please feel free to open an issue on GitHub. And discuss it on our Discord server.

Contributors

Support the Development by Donating

If you would like to support the development of Resume Matcher, you can do so by donating. Your contributions will help us keep the project alive and continue adding new features.

Platform	Link
GitHub
Buy Me a Coffee

Star History

EOS

Tue, 13 May 2025 15:30:26 +0800

Akkudoktor-EOS/EOS

Energy System Simulation and Optimization

This project provides a comprehensive solution for simulating and optimizing an energy system based on renewable energy sources. With a focus on photovoltaic (PV) systems, battery storage (batteries), load management (consumer requirements), heat pumps, electric vehicles, and consideration of electricity price data, this system enables forecasting and optimization of energy flow and costs over a specified period.

Documentation can be found at Akkudoktor-EOS.

Getting Involved

See CONTRIBUTING.md.

System requirements

Python >= 3.11, < 3.13
Architecture: amd64, aarch64 (armv8)
OS: Linux, Windows, macOS

Note: For Python 3.13 some dependencies (e.g. Pendulum) are not yet available on https://pypi.org and have to be manually compiled (a recent Rust installation is required).

Other architectures (e.g. armv6, armv7) are unsupported for now, because a multitude of dependencies are not available on https://piwheels.org and have to be built manually (a recent Rust installation and GCC are required, Python 3.11 is recommended).

Installation

The project requires Python 3.11 or newer. Docker images (amd64/aarch64) can be found at akkudoktor/eos.

Following sections describe how to locally start the EOS server on http://localhost:8503.

Run from source

Install dependencies in virtual environment:

Linux:

1
2
3

python -m venv .venv
.venv/bin/pip install -r requirements.txt
.venv/bin/pip install -e .

Windows:

python -m venv .venv
.venv\Scripts\Activate
.venv\Scripts\pip install -r requirements.txt
.venv\Scripts\pip install -e .

Finally, start the EOS server to access it at http://localhost:8503 (API docs at http://localhost:8503/docs):

Linux:

`1`	`.venv/bin/python src/akkudoktoreos/server/eos.py`

Windows:

`1`	`.venv\Scripts\python src/akkudoktoreos/server/eos.py`

Docker

Start EOS with following command to access it at http://localhost:8503 (API docs at http://localhost:8503/docs):

`1`	`docker compose up`

If you are running the EOS container on a system hosting multiple services, such as a Synology NAS, and want to allow external network access to EOS, please ensure that the default exported ports (8503, 8504) are available on the host. On Synology systems, these ports might already be in use (refer to this guide). If the ports are occupied, you will need to reconfigure the exported ports accordingly.

Configuration

This project uses the EOS.config.json file to manage configuration settings.

Default Configuration

A default configuration file default.config.json is provided. This file contains all the necessary configuration keys with their default values.

Custom Configuration

Users can specify a custom configuration directory by setting the environment variable EOS_DIR.

If the directory specified by EOS_DIR contains an existing config.json file, the application will use this configuration file.
If the EOS.config.json file does not exist in the specified directory, the default.config.json file will be copied to the directory as EOS.config.json.

Configuration Updates

If the configuration keys in the EOS.config.json file are missing or different from those in default.config.json, they will be automatically updated to match the default settings, ensuring that all required keys are present.

Classes and Functionalities

This project uses various classes to simulate and optimize the components of an energy system. Each class represents a specific aspect of the system, as described below:

Battery: Simulates a battery storage system, including capacity, state of charge, and now charge and discharge losses.
PVForecast: Provides forecast data for photovoltaic generation, based on weather data and historical generation data.
Load: Models the load requirements of a household or business, enabling the prediction of future energy demand.
Heatpump: Simulates a heat pump, including its energy consumption and efficiency under various operating conditions.
Strompreis: Provides information on electricity prices, enabling optimization of energy consumption and generation based on tariff information.
EMS: The Energy Management System (EMS) coordinates the interaction between the various components, performs optimization, and simulates the operation of the entire energy system.

These classes work together to enable a detailed simulation and optimization of the energy system. For each class, specific parameters and settings can be adjusted to test different scenarios and strategies.

Customization and Extension

Each class is designed to be easily customized and extended to integrate additional functions or improvements. For example, new methods can be added for more accurate modeling of PV system or battery behavior. Developers are invited to modify and extend the system according to their needs.

Server API

See the Swagger API documentation for detailed information: EOS OpenAPI Spec

Further resources

Installation guide (de)

less_slow.cpp

Mon, 21 Apr 2025 15:29:19 +0800

ashvardanian/less_slow.cpp

Playing Around Less Slow Coding Practices for C++, CUDA, and Assembly Code

The benchmarks in this repository don’t aim to cover every topic entirely, but they help form a mindset and intuition for performance-oriented software design. It also provides an example of using some non-STL but de facto standard libraries in C++, importing them via CMake and compiling from source. For higher-level abstractions and languages, check out less_slow.rs and less_slow.py. I needed many of these measurements to reconsider my own coding habits, but hopefully they’re helpful to others as well. Most of the code is organized in very long, ordered, and nested #pragma sections — not necessarily the preferred form for everyone.

Much of modern code suffers from common pitfalls — bugs, security vulnerabilities, and performance bottlenecks. University curricula and coding bootcamps tend to stick to traditional coding styles and standard features, rarely exposing the more fun, unusual, and potentially efficient design opportunities. This repository explores just that.

The code leverages C++20 and CUDA features and is designed primarily for GCC, Clang, and NVCC compilers on Linux, though it may work on other platforms. The topics range from basic micro-kernels executing in a few nanoseconds to more complex constructs involving parallel algorithms, coroutines, and polymorphism. Some of the highlights include:

100x cheaper random inputs?! Discover how input generation sometimes costs more than the algorithm.
1% error in trigonometry at 1/40 cost: Approximate STL functions like std::sin in just 3 lines of code.
4x faster lazy-logic with custom std::ranges and iterators!
Compiler optimizations beyond -O3: Learn about less obvious flags and techniques for another 2x speedup.
Multiplying matrices? Check how a 3x3x3 GEMM can be 70% slower than 4x4x4, despite 60% fewer ops.
Scaling AI? Measure the gap between theoretical ALU throughput and your BLAS.
How many if conditions are too many? Test your CPU’s branch predictor with just 10 lines of code.
Prefer recursion to iteration? Measure the depth at which your algorithm will SEGFAULT.
Why avoid exceptions? Take std::error_code or std::variant-like wrappers?
Scaling to many cores? Learn how to use OpenMP, Intel’s oneTBB, or your custom thread pool.
How to handle JSON avoiding memory allocations? Is it easier with C++ 20 or old-school C 99 tools?
How to properly use STL’s associative containers with custom keys and transparent comparators?
How to beat a hand-written parser with consteval RegEx engines?
Is the pointer size really 64 bits and how to exploit pointer-tagging?
How many packets is UDP dropping and how to serve web requests in io_uring from user-space?
Scatter and Gather for 50% faster vectorized disjoint memory operations.
Intel’s oneAPI vs Nvidia’s CCCL? What’s so special about <thrust> and <cub>?
CUDA C++, PTX Intermediate Representations, and SASS, and how do they differ from CPU code?
How to choose between intrinsics, inline asm, and separate .S files for your performance-critical code?
Tensor Cores & Memory differences on CPUs, and Volta, Ampere, Hopper, and Blackwell GPUs!
How coding FPGA differs from GPU and what is High-Level Synthesis, Verilog, and VHDL? 🔜 #36
What are Encrypted Enclaves and what’s the latency of Intel SGX, AMD SEV, and ARM Realm? 🔜 #31

To read, jump to the less_slow.cpp source file and read the code snippets and comments. Keep in mind, that most modern IDEs have a navigation bar to help you view and jump between #pragma region sections. Follow the instructions below to run the code in your environment and compare it to the comments as you read through the source.

Running the Benchmarks

The project aims to be compatible with GCC, Clang, and MSVC compilers on Linux, MacOS, and Windows. That said, to cover the broadest functionality, using GCC on Linux is recommended:

If you are on Windows, it’s recommended that you set up a Linux environment using WSL.
If you are on MacOS, consider using the non-native distribution of Clang from Homebrew or MacPorts.
If you are on Linux, make sure to install CMake and a recent version of GCC or Clang compilers to support C++20 features.

If you are familiar with C++ and want to review code and measurements as you read, you can clone the repository and execute the following commands.

git clone https://github.com/ashvardanian/less_slow.cpp.git # Clone the repository
cd less_slow.cpp                                            # Change the directory

pip install cmake --upgrade                                 # PyPI has a newer version of CMake
sudo apt-get install -y build-essential g++                 # Install default build tools
sudo apt-get install -y pkg-config liburing-dev             # Install liburing for kernel-bypass
sudo apt-get install -y libopenblas-base                    # Install numerics libraries

cmake -B build_release -D CMAKE_BUILD_TYPE=Release          # Generate the build files
cmake --build build_release --config Release                # Build the project
build_release/less_slow                                     # Run the benchmarks

The build will pull and compile several third-party dependencies from the source:

Google’s Benchmark is used for profiling.
Intel’s oneTBB is used as the Parallel STL backend.
Meta’s libunifex is used for senders & executors.
Eric Niebler’s range-v3 replaces std::ranges.
Victor Zverovich’s fmt replaces std::format.
Ash Vardanian’s StringZilla replaces std::string.
Hana Dusíková’s CTRE replaces std::regex.
Niels Lohmann’s json is used for JSON deserialization.
Yaoyuan Guo’s yyjson for faster JSON processing.
Google’s Abseil replaces STL’s associative containers.
Lewis Baker’s cppcoro implements C++20 coroutines.
Jens Axboe’s liburing to simplify Linux kernel-bypass.
Chris Kohlhoff’s ASIO as a networking TS extension.
Nvidia’s CCCL for GPU-accelerated algorithms.
Nvidia’s CUTLASS for GPU-accelerated Linear Algebra.

To build without Parallel STL, Intel TBB, and CUDA:

1
2

cmake -B build_release -D CMAKE_BUILD_TYPE=Release -D USE_INTEL_TBB=OFF -D USE_NVIDIA_CCCL=OFF
cmake --build build_release --config Release

To control the output or run specific benchmarks, use the following flags:

1
2
3

build_release/less_slow --benchmark_format=json             # Output in JSON format
build_release/less_slow --benchmark_out=results.json        # Save the results to a file instead of `stdout`
build_release/less_slow --benchmark_filter=std_sort         # Run only benchmarks containing `std_sort` in their name

To enhance stability and reproducibility, disable Simultaneous Multi-Threading (SMT) on your CPU and use the --benchmark_enable_random_interleaving=true flag, which shuffles and interleaves benchmarks as described here.

`1`	`build_release/less_slow --benchmark_enable_random_interleaving=true`

Google Benchmark supports User-Requested Performance Counters through libpmf. Note that collecting these may require sudo privileges.

`1`	`sudo build_release/less_slow --benchmark_enable_random_interleaving=true --benchmark_format=json --benchmark_perf_counters="CYCLES,INSTRUCTIONS"`

Alternatively, use the Linux perf tool for performance counter collection:

`1`	`sudo perf stat taskset 0xEFFFEFFFEFFFEFFFEFFFEFFFEFFFEFFF build_release/less_slow --benchmark_enable_random_interleaving=true --benchmark_filter=super_sort`

Project Structure

The primary file of this repository is clearly the less_slow.cpp C++ file with CPU-side code. Several other files for different hardware-specific optimizations are created:

$ tree .
.
├── CMakeLists.txt          # Build & assembly instructions for all files
├── less_slow.cpp           # Primary CPU-side benchmarking code with the majority of examples
├── less_slow_amd64.S       # Hand-written Assembly kernels for 64-bit x86 CPUs
├── less_slow_aarch64.S     # Hand-written Assembly kernels for 64-bit Arm CPUs
├── less_slow.cu            # CUDA C++ examples for parallel algorithms for Nvidia GPUs
├── less_slow_sm70.ptx      # Hand-written PTX IR kernels for Nvidia Volta GPUs
└── less_slow_sm90a.ptx     # Hand-written PTX IR kernels for Nvidia Hopper GPUs

Memes and References

Educational content without memes?! Come on!

Google Benchmark Functionality

This benchmark suite uses most of the features provided by Google Benchmark. If you write a lot of benchmarks and avoid going to the full User Guide, here is a condensed list of the most useful features:

->Args({x, y}) - Pass multiple arguments to parameterized benchmarks
BENCHMARK() - Register a basic benchmark function
BENCHMARK_CAPTURE() - Create variants of benchmarks with different captured values
Counter::kAvgThreads - Specify thread-averaged counters
DoNotOptimize() - Prevent compiler from optimizing away operations
ClobberMemory() - Force memory synchronization
->Complexity(oNLogN) - Specify and validate algorithmic complexity
->SetComplexityN(n) - Set input size for complexity calculations
->ComputeStatistics("max", ...) - Calculate custom statistics across runs
->Iterations(n) - Control exact number of iterations
->MinTime(n) - Set minimum benchmark duration
->MinWarmUpTime(n) - To warm up the data caches
->Name("...") - Assign custom benchmark names
->Range(start, end) - Profile for a range of input sizes
->RangeMultiplier(n) - Set multiplier between range values
->ReportAggregatesOnly() - Show only aggregated statistics
state.counters["name"] - Create custom performance counters
state.PauseTiming(), ResumeTiming() - Control timing measurement
state.SetBytesProcessed(n) - Record number of bytes processed
state.SkipWithError() - Skip benchmark with error message
->Threads(n) - Run benchmark with specified number of threads
->Unit(kMicrosecond) - Set time unit for reporting
->UseRealTime() - Measure real time instead of CPU time
->UseManualTime() - To feed custom timings for GPU and IO benchmarks