oneDNN
ciao
Our great sponsors
oneDNN | ciao | |
---|---|---|
5 | 3 | |
3,456 | 243 | |
2.5% | 4.1% | |
10.0 | 8.8 | |
6 days ago | about 2 months ago | |
C++ | Prolog | |
Apache License 2.0 | GNU Lesser General Public License v3.0 only |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
oneDNN
-
Blaze: A High Performance C++ Math library
If you are talking about non-small matrix multiplication in MKL, is now in opensource as a part of oneDNN. It literally has exactly the same code, as in MKL (you can see this by inspecting constants or doing high-precision benchmarks).
For small matmul there is libxsmm. It may take tremendous efforts make something faster than oneDNN and libxsmm, as jit-based approach of https://github.com/oneapi-src/oneDNN/blob/main/src/gpu/jit/g... is too flexible: if someone finds a better sequence, oneDNN can reuse it without major change of design.
But MKL is not limited to matmul, I understand it...
-
Arc & Deep Learning Frameworks
For completeness, it looks like this question was posted to the oneDNN GitHub repo and the response was to stay tune for updates.
- Keeping POWER relevant in the open source world
-
Intel oneDNN 2.5 released with experimental RISC-V support
From the release note of oneDNN v2.5:
-
Is gpu hardware tied to cpu ISA ?
Intel are trying to support their oneAPI compute framework on Arm and IBM POWER and z/Architecture (s390x) but since they ever released only a single discrete GPU with the Xe architecture it's unclear whether they'll support Xe GPU compute on e.g. ARM https://github.com/oneapi-src/oneDNN
ciao
- PHP: Prolog Home Page
-
An embeddable Prolog scripting language for Go
Some Prolog systems (like Ciao Prolog https://github.com/ciao-lang/ciao/blob/master/core/lib/forei...) implement bidirectional foreign interfaces. Once you have C bindings it is easy to write bindings from Rust, C++, or any other language (that can interoperate with C). I give here some details about Ciao because this is the system I know better but it should be similar for other popular Prolog implementations.
The tradeoffs depend on the complexity of the Prolog code and your needs for performance and features: pure LP, Prolog (search+unification+cut), garbage collection, dynamic database updates, constraint domains, etc. The Ciao Prolog engine is around 300-400KB. Adding a few libraries, compiler, etc. it goes to 2MB. Naive Prolog systems can be one order of magnitude smaller at the cost of sacrificing ISO compatibility, performance, etc. Note that "performance" can be very misleading. Some Prolog programs may run particularly fast in some Prolog system and very badly in others.
- Keeping POWER relevant in the open source world
What are some alternatives?
oneMKL - oneAPI Math Kernel Library (oneMKL) Interfaces
prolog - The only reasonable scripting engine for Go.
CTranslate2 - Fast inference engine for Transformer models
Prolog-to-List-Prolog - Converts Prolog algorithms to List Prolog algorithms
oneDPL - oneAPI DPC++ Library (oneDPL) https://software.intel.com/content/www/us/en/develop/tools/oneapi/components/dpc-library.html
trealla-js - Trealla Prolog for the web
highway - Highway - A Modern Javascript Transitions Manager
go - Trealla Prolog embedded in Go using WASM
asmjit - Low-latency machine code generation
librealsense - IntelĀ® RealSenseā¢ SDK
php - Prolog Home Page