GithubHelp home page GithubHelp logo

zzhou612 / aquatope Goto Github PK

View Code? Open in Web Editor NEW
15.0 15.0 4.0 10.23 MB

AQUATOPE: QoS-and-Uncertainty-Aware Resource Management for Multi-Stage Serverless Workflows (ASPLOS'23)

License: Apache License 2.0

Python 90.54% Dockerfile 8.00% Shell 1.05% Makefile 0.41%

aquatope's People

Contributors

zzhou612 avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar

aquatope's Issues

Details on Aquatope and OpenWhisk Deployment

@zzhou612 Hello, amazing work! I am trying to reproduce the work in your paper. Could you provide more details on how you deployed Aquatope and OpenWhisk for your paper results? Specifically, the following information would be very helpful:

  • Arguments for --n_input_steps, --n_output_steps, etc. when training train_lstm_encoder_decoder.py.
  • Arguments for --n_input_steps', '--n_output_steps, etc. when training train_prediction_network.py
  • Arguments for deploying the container pool scheduler
  • Arguments for deploying the container resource manager

Finally, to deploy OpenWhisk, I have a few questions:

  1. How did you deploy OpenWhisk? Ansible?
  2. Other than merging the two mentioned PRs in the README, were there any other OpenWhisk changes that had to be made? Were the two PRs complete and usable right out the gate?

Could we get a detailed step-by-step deployment guidance?

Hi Zhuangzhuang,

Very impressive work!

But after seeing your updated read me section, I still don’t know exactly how to configure and reproduce the experiment. I mainly use kubernetes so don't know much about Openwhisk. If possible, could you provide a simple step-by-step instructions? According to my current understanding, we need to install the necessary libraries first and then run the training script. But I'm not very clear on how to start the corresponding components and how to test them later.

Thank you very much for any help!

Details on deployment

Hi Zhuangzhuang,

I'm trying to reproduce your amazing work. I have trained models with the help of your deployment instructions, but failed to run the scheduler. I have some questions here.

How did you generate your workload or how can I generate my workload with you scripts? Could you please write a step-by-step guidance?

How can I create a workflow config file? I have used your scripts in utils/init_config but config["OpenWhisk"]["GET_RUNTIME_URL"] and config["OpenWhisk"]["CONTROLLER_CHANGE_RUNTIME_URL"] are missing.

Thank you very much for any help!

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.