JacobZ

How is Modular Democratizing AI Compute? (Democratizing AI Compute, Part 11)

Mon, 13 Oct 2025 20:00:00 +0800

Repost Notes

This blog series is reposted from Modular’s official blog, written by the creator of LLVM, Chris Lattner.

Source: How is Modular Democratizing AI Compute? (Democratizing AI Compute, Part 11)
Publish date: June 20, 2025

The original blog was parsed with Jina Reader.

Go to webpage to read full text.

Modular’s bet to break out of the Matrix (Democratizing AI Compute, Part 10)

Mon, 13 Oct 2025 19:00:00 +0800

Repost Notes

This blog series is reposted from Modular’s official blog, written by the creator of LLVM, Chris Lattner.

Source: Modular’s bet to break out of the Matrix (Democratizing AI Compute, Part 10)
Publish date: May 8, 2025

The original blog was parsed with Jina Reader.

Go to webpage to read full text.

Why do HW companies struggle to build AI software? (Democratizing AI Compute, Part 9)

Mon, 13 Oct 2025 18:00:00 +0800

Repost Notes

This blog series is reposted from Modular’s official blog, written by the creator of LLVM, Chris Lattner.

Source: Why do HW companies struggle to build AI software? (Democratizing AI Compute, Part 9)
Publish date: April 22, 2025

The original blog was parsed with Jina Reader.

Go to webpage to read full text.

What about the MLIR compiler infrastructure? (Democratizing AI Compute, Part 8)

Mon, 13 Oct 2025 17:00:00 +0800

Repost Notes

This blog series is reposted from Modular’s official blog, written by the creator of LLVM, Chris Lattner.

Source: What about the MLIR compiler infrastructure? (Democratizing AI Compute, Part 8)
Publish date: April 8, 2025

The original blog was parsed with Jina Reader.

Go to webpage to read full text.

What about Triton and Python eDSLs? (Democratizing AI Compute, Part 7)

Mon, 13 Oct 2025 16:00:00 +0800

Repost Notes

This blog series is reposted from Modular’s official blog, written by the creator of LLVM, Chris Lattner.

Source: What about Triton and Python eDSLs? (Democratizing AI Compute, Part 7)
Publish date: March 26, 2025

The original blog was parsed with Jina Reader.

Go to webpage to read full text.

What about TVM, XLA, and AI compilers? (Democratizing AI Compute, Part 6)

Mon, 13 Oct 2025 15:00:00 +0800

Repost Notes

This blog series is reposted from Modular’s official blog, written by the creator of LLVM, Chris Lattner.

Source: What about TVM, XLA, and AI compilers? (Democratizing AI Compute, Part 6)
Publish date: March 12, 2025

The original blog was parsed with Jina Reader.

Go to webpage to read full text.

What about OpenCL and CUDA C++ alternatives? (Democratizing AI Compute, Part 5)

Mon, 13 Oct 2025 14:00:00 +0800

Repost Notes

This blog series is reposted from Modular’s official blog, written by the creator of LLVM, Chris Lattner.

Source: What about OpenCL and CUDA C++ alternatives? (Democratizing AI Compute, Part 5)
Publish date: March 5, 2025

The original blog was parsed with Jina Reader.

Go to webpage to read full text.

CUDA is the incumbent, but is it any good? (Democratizing AI Compute, Part 4)

Mon, 13 Oct 2025 13:00:00 +0800

Repost Notes

This blog series is reposted from Modular’s official blog, written by the creator of LLVM, Chris Lattner.

Source: CUDA is the incumbent, but is it any good? (Democratizing AI Compute, Part 4)
Publish date: February 20, 2025

The original blog was parsed with Jina Reader.

Go to webpage to read full text.

How did CUDA succeed? (Democratizing AI Compute, Part 3)

Mon, 13 Oct 2025 12:00:00 +0800

Repost Notes

This blog series is reposted from Modular’s official blog, written by the creator of LLVM, Chris Lattner.

Source: How did CUDA succeed? (Democratizing AI Compute, Part 3)
Publish date: February 12, 2025

The original blog was parsed with Jina Reader.

Go to webpage to read full text.

What exactly is "CUDA"? (Democratizing AI Compute, Part 2)

Mon, 13 Oct 2025 11:00:00 +0800

Repost Notes

This blog series is reposted from Modular’s official blog, written by the creator of LLVM, Chris Lattner.

Source: What exactly is “CUDA”? (Democratizing AI Compute, Part 2)
Publish date: Feburary 5, 2025

The original blog was parsed with Jina Reader.

Go to webpage to read full text.

DeepSeek’s Impact on AI (Democratizing AI Compute, Part 1)

Mon, 13 Oct 2025 10:00:00 +0800

Repost Notes

This blog series is reposted from Modular’s official blog, written by the creator of LLVM, Chris Lattner.

Source: DeepSeek’s Impact on AI (Democratizing AI Compute, Part 1)
Publish date: January 30, 2025

The original blog was parsed with Jina Reader.

Go to webpage to read full text.

Brief Tutorial for Mersenne Prime Test Softwares

Wed, 22 Feb 2023 00:00:00 +0000

After playing with linux OSes recently, I also tried to figure out other ways to contribute to GIMPS. In this blog, I will summarize the softwares recommended for the GIMPS project and their usage. GIMPS relies on the PrimeNet for tasks distribution and collection. Currently there are three kinds of tasks for searching new Mersenne primes will be distributed: TrialT FactoringF、P-1PM1 factorization and Probable PrimePRP test。Therefore, I will mainly cover the usage and recommended strategy in this blog.

Go to webpage to read full text.

The Pitfall Of Long Double

Sun, 24 Jul 2022 00:00:00 +0000

Repost Notes

This article is a repost, originally written by David on the Prelert’s website. The original article is now only accessible on WebArchive.

I found this article when I’m searching for the reason why Rust doesn’t have a corresponding type for long double in C/C++, which have caused some interoperability issues (see here and here). On the contrary, the languages Zig and the newly born Carbon both support f16 and f128 types (Zig also supports f80 and Carbon also supports bfloat16). But that’s not suprising because they all aim to provide max interoperability with C/C++. This article might explain some of the reason why Rust doesn’t support float types with higher precision.

Go to webpage to read full text.

Select Modern Programming Languages by One-line comments

Thu, 05 Aug 2021 00:00:00 +0000

In recent years, Many novel programming languages have emergerd, and new concepts continued to appear. I always wanted to learn about various new programming languages, and it would be nice to master one more language if it is prospective. Therefore, this article summarizes my understanding of some popular modern languages and their various characteristics. If you are also interested in trying a new language, then I hope this article can help you~

Go to webpage to read full text.

Notes for Algebra Basics

Sat, 27 Jun 2020 00:00:00 +0000

In this note, $\{x_i\}^b_a$ denotes set $\{x_a, x_{a+1}, \ldots, x_b\}$

Algebraic Structures

Operation

Definition: an (binary, closed) operation $\ast$ on a set $S$ is a mapping of $S\times S\to S$
Commutative: $x\ast y=y\ast x,\;\forall x,y\in S$
Associative: $(x\ast y)\ast z=x\ast (y\ast z),\;\forall x,y,z\in S$

Group

Definition: a group is a pair $(\mathcal{S},\ast)$ with following axioms
1. $\ast$ is associative on $\mathcal{S}$
2. (Identity element) $\exists e\in \mathcal{S}\text{ s.t. }x\ast e=e\ast x=x,\;\forall x\in \mathcal{S}$
3. (Inverse element) $\forall x\in \mathcal{S}, \exists x’ \in \mathcal{S}\text{ s.t. }x\ast x’=x’\ast x=e$
Abelian: a group is called abelian group if $\ast$ is also commutative

Ring

Definition: a ring is a triplet $(\mathcal{R},+,\ast)$ consisting of a set of scalars $\mathcal{R}$ and two operators + and $\ast$ with following axioms
1. $(\mathcal{R},+)$ is an abelian group with identity denoted $0$
2. $\forall a,b,c \in \mathcal{R}\text{ s.t. }a\ast(b\ast c) = (a\ast b)\ast c$
3. $\exists 1\in\mathcal{R}, \forall a\in\mathcal{R}\text{ s.t. }a\cdot 1=a$
4. $\ast$ is distributive over $+$

Field

Definition: a field $(\mathcal{F},+,\ast)$ is a ring where $(\mathcal{F}\backslash\{0\},\ast)$ is also an abelian group.

Difference from ring to field is that $\ast$ need to be commutative and have a multiplicative inverse

Go to webpage to read full text.