31
Paving the way to efficient architectures: StripedHyena-7B
StripedHyena-7B is an open-source model built beyond the Transformer architecture, matching Llama-2 and Mistral 7B while running faster on long sequences.
New
The latest blog posts to take off on Hacker News. 555 posts.
32
33
For processing strings, streams in C++ can be slow
The C++ library has long been organized around stream classes, at least when it comes to reading and parsing strings. But streams can be surprisingly slow. For example, if you want to parse numbers, then this C++ routine is close to being the worst possible choice for performance: std::stringstream in(mystring)…
34
Virtual Machine as a core Android Primitive
News and insights on the Android platform, developer tools, and events.
35
36
37
A minimum complete tutorial of Linux ext4 file system (2017)
This is a complete tutorial for understanding the basics of Linux ext4 file system.
38
Buggy animation in Atlassian Bitbucket is wasting half a CPU core at all times
I was running some benchmarks the other day. I usually have a lot of things running (Teams, Chrome, IDE, etc…) but I have to close everything to run some tight benchmarks. When I am done, one of th…
39
The Intel 386 processor die: the clock circuit
Processors are driven by a clock, which controls the timing of each step inside the chip. In this blog post, I'll examine the clock-generati...
40
XCurl
It is often said that Imitation is the Sincerest Form of Flattery. Also, remember libcrurl? That was the name of the thing Google once planned to do: reimplement the libcurl API on top of their Chrome networking library. Flattery or not, it never went anywhere. The other day I received an email asking me about ……
41
Rust std fs slower than Python? No, it's hardware
Achieving Data Freedom Through Open Source and Rust
42
43
44
Eclipse ThreadX
TL;DR – Get Engaged! Contact us to learn how to become part of the Eclipse ThreadX Interest Group Contribute to the project and join the ThreadX developer mailing list Stay connected by joini…
45
Designing a SIMD Algorithm from Scratch
Another explainer on a fun, esoteric topic: optimizing code with SIMD (single instruction multiple data, also sometimes called vectorization ). Designing a good, fast, portable SIMD algorithm is not a simple matter and requires thinking a little bit like a circuit designer.
46
Evaluating M3 Pro CPU Cores: General Performance
Using assembly language test loops to understand the differences between M1 Pro and M3 Pro CPU cores casts new light on their differences.
47
C++ Memory Model: Migrating from x86 to ARM (2021)
Explore the C++ memory model and learn about migrating from x86 to ARM architectures. Enhance your understanding of memory management in C++.
48
Windows NT: Peeking into the Cradle
A review of "Showstopper!" by G. Pascal Zachary and my own reflections
49
Retries – An interactive study of request retry methods
An interactive guide to retrying requests: why naive retries overload a failing service, and how exponential backoff and jitter fix it.
50
Writing a Debugger from Scratch: Stacks
(New to this series? Consider starting from part 1) At the end of the last post, DbgRs could finally set breakpoints. That was the last really critical feature needed for controlling execution of a program. Now that it can stop on an arbitrary function or code location, the next step is to find more information about…
51
Fast Llama 2 on CPUs with Sparse Fine-Tuning and DeepSparse
From live events to hands-on product demos to deep technical research, see what we're doing with AI at Red Hat.
52
What has changed in CPU cores in M3 chips?
Little seems to have changed in CPU cores in M3 chips, if you read reviews. Dig a bit deeper and there are major changes, as explained here.
53
FreeBSD 14.0-Release
The FreeBSD Release Engineering Team is pleased to announce the availability of FreeBSD 14.0-RELEASE. This is the first release from the stable/14 branch.
54
Increasing QUIC and UDP Throughput over Tailscale
Recent optimizations in Tailscale improve UDP throughput, boosting performance for QUIC, HTTP/3, and other UDP-based applications on Linux.
55
Setenv Is Not Thread Safe and C Doesn't Want to Fix It
You can't safely use the C setenv() or unsetenv() functions in a program that uses threads. Those functions modify global state, and can cause other threads calling getenv() to crash. This also causes crashes in other languages that use those C standard library functions, such as Go's os.Setenv ( Go issue ) and Rust's…
56
57
Operating on a minimal two-core Postgres instance: Query optimization insights
In an era where the costs for cloud services continue to soar, I want to outline how our data team has successfully managed to run the infrastructure of our energy trading company within the first year using only two PostgreSQL database instances - each equipped with just two CPU cores and four GB of RAM. Despite this…
58
Curl on 100 Operating Systems
In a recent pull-request for curl, I clarified to the contributor that their change would only be accepted and merged into curl's git code repository if they made sure that the change was done in a way so that it did not break (testing) for and on legacy platforms. In that thread, I could almost … Continue reading…
59
The Architecture of Serverless Data Systems
I recently blogged about why I believe the future of cloud data services is large-scale and multi-tenant, citing, among others, S3. “ Top tier SaaS services like S3 are able to deliver amazing simplicity, reliability, durability, scalability, and low price because their technologies are s
60
Reptar
We have a CPU mystery! We found a way to cause some processors to enter a glitch state where the normal rules don’t apply, but what does that mean…?