FlowQ 0.0.3

Your Friendly Neighbourhood Distributed Computing System

FlowQ

Your Friendly Neighbourhood Distributed Computing System

FlowQ was created for the goal of making distributed computing free, simple and easier.

Features of FlowQ:

• Effortless Setup: Ditch the complicated configurations! FlowQ runs right out of the box, no ssh headaches or pre-installation required.
• Simple and Secure Connection: Leverages the Hack.Chat platform to establish secure, base-64 encrypted and anonymous connections with your computing cluster.
• Temporary Storage: Need a place to store input and output files? FlowQ utilizes FileBin for convenient temporary storage.
• Parallel Powerhouse: FlowQ unleashes the true potential of your network by executing tasks in parallel across your machines(with multi-threading), significantly boosting your processing speed.

I don't have any other computing devices, and I don't want to spend money....

• Supercharge your cluster in seconds! FlowQ lets you seamlessly add new machines with Python. Just 2 lines of command, and you've got a processing powerhouse. FlowQ makes scaling effortless.

pip install FlowQ
python -m FlowQ.cluster -c <your-channel-name>

• You can run these commands in your Google Colab Instances or any other computer, for scaling your cluster with ease.

Client Usage

• You can set up your client, with simple FlowQlient Class!

from FlowQ.client import FlowQlient
flow = FlowQlient(channel="<your-channel-name>")
flow.connect(name="<your-user-name>")

@flow.task
def alpha(x):
    ## Import all the needed modules inside the function
    import requests
    url = "https://uselessfacts.jsph.pl/api/v2/facts/random"
    return requests.get(url).json()["text"], x

output = flow.get([alpha(i) for i in range(6)])

⚡Note⚡: Please initialize the cluster before running the Client code(This will be fixed in future updates)

⚡Note⚡: While executing the client in a jupyter notebook, don't forget to enable nest_asyncio first by:

!pip install nest_asyncio
import nest_asyncio
nest_asyncio.apply()