MLProxy: SLA-Aware Reverse Proxy for Machine Learning Workloads

Codes, packages, experimentation results, and other artifacts for our SLA-Aware reverse proxy paper. Using the proposed reverse proxy, developers can reduce the cost of deployment on serverless computing platforms and improve the performance while making sure SLA objectives are met in the process. For more information, please refer to our paper.

Artifacts

Here is a list of artifacts for the proposed study:

• Experiment Code for Deployment and Runtime
• MLProxy's Code and Dockerfile for Reproducibility
• Traces used for our experimentations
• Machine Learning Workloads Used For Our Experimentations

Requirements

• Python 3.7+
• PIP
• Node 14.18.2+
• Docker

License

Unless otherwise specified:

MIT (c) 2020 Nima Mahmoudi & Hamzeh Khazaei

Citation

You can find the paper with details of the proposed model in PACS lab website. You can use the following bibtex entry:

coming soon...