Git Product home page Git Product logo

snax_rebase's Introduction

CI License

SNAX Cluster

This repository hosts the hardware and software for the SNAX cluster and its generator. SNAX is a high-efficiency compute cluster platform designed for computation- and data- intensive artificial intelligence workloads. It provides a standard SNAX shell for conveniently integrating versatile accelerators. The SNAX shell leverages CSR manager for programming the accelerators in the control plane and data streamers for data access in the data plane.

The figure below shows an architectural overview of the SNAX cluster.

image

SNAX supports several design-time and run-time configurations to support a variety of accelerators. For example, some design-time configurations include customizing the memory sizes, the interconnect structure connecting the accelerators to memory, the number of Snitch cores controlling accelerators, and so much more. For run-time configurations, we provide data streamers and reshufflers to aid accelerators in organizing data layouts in memory and managing data access patterns.

Getting Started

To get started, check out the getting started guide.

We provide a detailed tutorial on integrating an accelerator into the SNAX cluster platform.

Content

What can you expect to find in this repository?

  • The RISC-V Snitch integer core. This can be useful stand-alone if you are just interested in re-using the core for your project, e.g., as a tiny control core or you want to make a peripheral smart. The sky is the limit.

  • The SNAX cluster. A highly configurable cluster provides a standard and clean accelerator integration interface.

  • A CSR Manager is included to support a standardized control and management of registers through CSR instructions.

  • The Data Streamer streamlines the data access patterns for the accelerator.

  • Versatile accelerator examples, including a Chisel-based GEMM accelerator generator, a rescale SIMD accelerator generator, and a Data Reshuffler for data layout transformation, are provided for reference.

  • Software example applications for the SNAX cluster and corresponding accelerators.

  • RTL simulation environments for Verilator, Questa Advanced Simulator, and VCS.

Directory Structure

The project is organized as a monolithic repository. Both hardware and software are co-located. Standalone accelerators can have their own repositories and can be imported here using Bender.

The file tree is visualized as follows:

├── hw
│   ├── chisel
│   │   ├── csr_manager
│   │   └── streamer
│   ├── snax_accelerator_1
│   ├── snax_accelerator_2
│   ├── snitch_stuff
│   └── templates
├── sw
├── target
│   └── snitch_cluster
│       ├── config
│       ├── generated
│       └── sw
│           ├── apps
│           │   ├── snax_system_1
│           │   └── snax_system_2
│           └── snax_lib 
│               ├── snax_system_1
│               └── snax_system_2
└── util
    ├── clustergen
    └── wrappergen

The top-level is structured as follows:

  • docs: Documentation of the generator and software. Contains additional user guides.
  • hw: All hardware IP components. The source files are either specified by SystemVerilog, Chisel, or a template to generate these files.
  • sw: Hardware independent software, libraries, runtimes, etc.
  • target: Contains the testbench setup, cluster configuration specific hardware and software, libraries, runtimes, etc.
  • util: Utility and helper scripts.

Acknowledgement

SNAX is an extension of the original Snitch Cluster framework. SNAX got its name as "SNitch Acceleration eXtension".

snax_rebase's People

Contributors

zarubaf avatar colluca avatar paulsc96 avatar rgantonio avatar fabianschuiki avatar huettern avatar giannap avatar xiaoling-yi avatar fischeti avatar jossevandelm avatar jorendumoulin avatar viv-eth avatar samuelriedel avatar thommythomaso avatar niwis avatar lucabertaccini avatar cyrilkoe avatar zero9178 avatar micprog avatar raragyay avatar and-ivanov avatar flaviens avatar haugoug avatar suehtamacv avatar rogerbarton avatar iveanex avatar otoomey avatar

Watchers

 avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.