enduku

36 ms·

31.
A single-file C allocator with explicit heaps and tuning knobs(github.com)
57 points by enduku 7mo ago|44 comments
32.
by enduku 7mo ago
I wrote this because I wanted more explicit control over heaps when building different subsystems in C. Standard options like jemalloc and mimalloc are incredibly fast, but they act as black boxes. You can't easily cap a parser's
33.
by enduku 2y ago
https://en.algorithmica.org/
34.
How branch prediction works in CPUs [video](youtube.com)
2 points by enduku 2y ago|0 comments
35.
Not Just Scale – Why thinking in distributed systems is important(brooker.co.za)
14 points by enduku 2y ago|0 comments
36.
by enduku 3y ago
Mirdin[0] can be helpful too. [0] https://self-service.mirdin.com/products/advanced-software-d...
37.
TinyChat: Large Language Model on the Edge(hanlab.mit.edu)
2 points by enduku 3y ago|1 comments
38.
by enduku 3y ago
TinyChat is an efficient, lightweight, Python-native serving framework for 4-bit LLMs by AWQ. It delivers 2.3x generation speed up on RTX4090. Code: https://github.com/mit-han-lab/llm-awq/tree/main/tinych
39.
by enduku 3y ago
I find myself using J [0] for such type of ideation/programming exercises. Klong/K/APL also lend to similar productivity. [0] https://www.hillelwayne.com/handwriting-j/
40.
by enduku 3y ago
LibUI for C [0] [0] https://github.com/andlabs/libui
41.
by enduku 3y ago
Kagi search is brilliant in this aspect. There used to be a similar service [0] not too long ago that blocked many spam sites from the search results. [0] http://millionshort.com/
42.
by enduku 3y ago
Mine is https://densebit.com I have a tendency to jot down brief thoughts or raw ideas :-)
43.
The Hyena code is able to handle data that makes GPT models throw OOM error(zdnet.com)
1 points by enduku 3y ago|0 comments
44.
Gio Wiederhold the database design pioneer passes away(en.wikipedia.org)
3 points by enduku 4y ago|0 comments
45.
50 years of C, the good, the bad and the ugly [video](streaming.media.ccc.de)
156 points by enduku 4y ago|254 comments
46.
by enduku 4y ago
The code correctness issue is certainly a big problem -- it is simply not enough to get 90% correct. The real world problems often lie in the remaining 10% edge cases. I have a very different take on how AI can come up with a correct by con
47.
Kawipiko – fast static HTTP server in Go(github.com)
68 points by enduku 4y ago|22 comments
48.
by enduku 4y ago
A new way of looking at supervised learning: as generalizations of variational calculus.
49.
Do Ideas Have Shape?(arxiv.org)
2 points by enduku 4y ago|1 comments
50.
by enduku 4y ago
REBOL Parse maybe?
51.
by enduku 4y ago
Prescriptive analytics for Pharma.
52.
by enduku 4y ago
Yes. Kdb+ and JD come to mind whose primary data structures are multidimensional arrays.
53.
by enduku 5y ago
Great work, congrats! Another app Beeline Reader[0] seems a lot similar to this. It was launched many years ago I suppose. All the best with the product! [0] https://www.beelinereader.com/
54.
by enduku 5y ago
red-lang isn't bad. Golang. Terra. Nim. Forth. Pascal. Zig. Rust.
55.
The Design and Topology of an Algorithm(densebit.com)
1 points by enduku 5y ago|0 comments
56.
by enduku 5y ago
In addition to wonderful responses already posted, I'd like to add one more: Learn Forth. Learn Lisp. Learn APL. No need to be good at these; just enough to learn their programming models. Imperative, functional, OOP, or any other, it
57.
Some C and C++ refactoring tips/pitfalls(software.intel.com)
2 points by enduku 5y ago|1 comments
58.
by enduku 5y ago
A few useful common refactoring, programming and other C and C++ static analysis tips.
59.
by enduku 5y ago
Based on the number of CPU cores on the machine, Shenandoah sets up its threads. JVM will see only one CPU core if the number of CPU shares for the container is not set explicitly[0]. In that case, Shenandoah will use only single parallel t
60.
by enduku 5y ago
One neat feature of Shenandoah is once the application shuts down, it prints into the GC log a table depicting how much time each phase of the GC cycle took. This data can be crucial to optimize the GC pause time further: - Removal of weak

More ›