Rpi Cluster Notes

Viewed 5 times

I am new to Raspberry Pi’s (as well as this stackexchange) and would like to use them to learn how to build a small cluster. (Currently using this guide: https://medium.com/@glmdev/building-a-raspberry-pi-cluster-784f0df9afbd)

I am a mathematics graduate student and so I am trying to gauge whether or not this project will actually be of use for some of my work. I would want to use a cluster for doing hundreds of millions of parallel linear algebra computations in R, Python and SageMath.

I would like to get an idea of the computational power of just one Raspberry Pi 4 Model B 2019 Quad Core 64 Bit WiFi Bluetooth (4GB). How would this compare to a cluster with something like Intel Core i3 processors, for example. Any advice and suggestions would be greatly appreciated! Thank you for your time!

Perhaps this post is relevant: Cost-effectiveness of Pi cluster

 New contributor
  • Hi @bark, Welcome and nice to meet you. I skimmed Parts 1, 2, and 3, and I impressed. Let me give some quick comments. (1) The project is very well presented for newbies. (2) Of course it is an advanced topic in parallel/ concurrent computing, but using Open MPI and python make it not scary at all for newbies. (3) Of course newbies need to understand that it is not at all a weekend project. If you only know blinking a LED using python, you might need one year’s hobbyist time to complete the project. – tlfong01 14 mins ago   
  • (4) For sharing storage, newbies might like to first try Samba and PureFTD sharing files with Win10. All Rpi might be a bit hard for newbies to develop and troubleshoot (Yes, I am a WinPC guy and sort of linux newbie.) (5) For mass storage, I am using 1/2Gb USB SSD and found it newbie friendly than HDD, because tthey just look like USB drives. (6) The project started 2018. Rpi4B appeared 2019 and a couple of things got greatly improved, including USB3, 1Gb Ethernet, both of which help making the cluster a couple of time more powerful and faster. (7) Python MPI is of course good, … – tlfong01 9 mins ago   
  • Pyhon 3.7x already it self has mutl-processing (more newbie friendly than old style multi-threading). Of course python is ideal for google style concurrent processing, including TensorFlow which almost always uses python as examples. (8) We need to know this is kind of educational/learning project. We need to appreciate that we are not make use of GPU, and USB3, though so much faster, is still a severe communication bottleneck, limiting performance of parallel processing. (9) I have been reading similar Rpi based cluster projects these years, for learning, I would give 5 marks out of 5. – tlfong01 7 mins ago   
  • Just thinking aloud, sorry for the typos. I have 4 Rpi4B 1/2 Gb, and have been playing with python 3.7 multiprocessing, 1/2TB USB SSD, Win10/Rpi3B+ Samba/PureFTD, etc. I also have a 4 years old PC with CORE i5, 6GB RAM, Nvidia, etc. As I said, Rpi USB3 slow communication and no GPU is the fatal weakness, Rpi x 4 can’t compare with single evil Wintel PC! 🙂 – tlfong01 just now   Edit


Categories: Uncategorized

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

<span>%d</span> bloggers like this: