https://github.com/aratz-lasa/dgraph_scaler

Distributed Graph Scaler
https://github.com/aratz-lasa/dgraph_scaler

Last synced: 13 days ago
JSON representation

Distributed Graph Scaler

• Host: GitHub
• URL: https://github.com/aratz-lasa/dgraph_scaler
• Owner: aratz-lasa
• License: mit
• Created: 2020-11-16T19:27:57.000Z (about 4 years ago)
• Default Branch: master
• Last Pushed: 2020-12-16T22:32:32.000Z (about 4 years ago)
• Last Synced: 2024-11-07T15:53:48.529Z (2 months ago)
• Language: Python
• Size: 9.94 MB
• Stars: 1
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE.md

Awesome Lists containing this project

README

        
# dgraph_scaler

Graph dataset scaler that receives an input graph and it scales up or down by any arbitrary factor.

## System architecture

	

The system presents a flat architecture, where all the nodes are homogeneous in configuration and resources. 

There is no hierarchy. This way, unnecessary complexities in the programming model are avoided. 

Any node can communicate with any other node. For communication the MPI library is used. 

This way, the nodes do not need to take care of naming, discovery or higher-level multiplexing abstractions and logic.

![alt text](architecture.png "System srchitecture")

## Usage

In order to use the program with MPI the following command is used:

```console

foo@bar:~$ mpirun -np 2 python3.5 main.py datasets/graph.txt  samples/graph 6.5 -v

```

There are many scaling options. You can check them by running:

```console

foo@bar:~$ python3.5 main.py --help

```