View on GitHub

Graphitti

A project to facilitate construction of high-performance neural simulations

1.1 Introduction

1.1.1 What is Graphitti?

Graphitti is an open-source high-performance graph-based event simulator that is intended to aid scientists and researchers by providing pre-built code that can easily be modified to fit different models.

Graphitti runs on both a CPU & GPU. The multi-threaded (GPU) version uses NVIDIA’s CUDA libraries and can provide a potential speedup of up to a twentieth of the original runtime. The single-threaded (CPU) simulator compiles on a machine without CUDA libraries.

1.1.2 What is Graphitti for?

Graphitti is a framework for quickly implementing GPU-speed-up graph-based networks. If you have a mathematical model for real world events, Graphitti will provide you with a way of getting that network implemented in code and running on CPUs or GPUs.

In sum, Graphitti is designed to do the following:

  1. Simulate graph-based networks with built-in models. For example, the neuroscience simulator can do this with the leaky-integrate-and-fire neuron and izhikevich neuron.
  2. Provide a framework to make it easier to code models of your own.
  3. Make these models FAST - provide code to make it easy to migrate models to single or multiple GPUs, providing a huge speedup from single or multi-threaded host-side models.

1.1.3 Why do we need Graphitti?

The initial principles that we are basing Graphitti on are as follows:

1.1.4 Who is the Graphitti Audience?


» Next: 1.2 Installation


« Go back to User Documentation page


« Go back to Graphitti home page