This is part two of the series that shows you how to make a cluster computer using Raspberry Pi's!
You can watch Part 1 here.
davisde | May 9, 2014
[venera_framed_image content_type="video" css_animation="appear" frame_type="browser" slider_engine="flexslider" video_link="" browser_url="https://www.youtube.com/watch?v=lHmFRlETTcQ"]
This is part two of the series that shows you how to make a cluster computer using Raspberry Pi's!
You can watch Part 1 here.
TOTAL COST: ~$100.00
[tabby title="Terminal Commands"]
This post currently has 4 responses.
davisde | December 11, 2021
davisde | March 13, 2021
davisde | February 10, 2018
davisde | January 13, 2018
davisde | May 27, 2017
davisde | May 2, 2017
davisde | April 15, 2017
davisde | March 18, 2017
pacmandu
September 9, 2015 at 4:20 am
Your python script isn’t actually using MPI. You need to actually import MPI library and use their API’s for distributing tasks. All you are doing is running the same script on the Pi’s and not getting any more speed than just one Pi.
pacmandu
September 9, 2015 at 4:21 am
from mpi4py import MPI
pacmandu
September 9, 2015 at 4:27 am
Here is the actual code for an MPI md5 python cracker. https://gist.githubusercontent.com/zooty/110cb97dcbeb17210765/raw/b83a30d382869e35b33f5f39be97d73ee1071e70/cracker.py
pacmandu
September 9, 2015 at 4:44 am
Sorry but I still did not link to the correct MPI python code but you need to setup your code to know to send each word from the dictionary to one of the nodes in your cluster to hash and compare and then when one of the nodes reports back that it has a match you stop execution across all nodes. That unfortunately is not what your code is doing. When I build my cluster I will post a correct version that uses MPI to distribute the MD5 crack process.