Read the original document by opening this link in a new tab.
Table of Contents
1. Introduction
2. Motivating Example
3. Proposed Solution
4. Feasibility
5. Related Work
6. Conclusion
Acknowledgments
References
Summary
The document discusses the challenges and requirements of real-time machine learning applications, proposing a new distributed execution framework to meet these demands. It introduces a programming model and architecture to address latency, throughput, dynamic task creation, and fault tolerance. Results show promising performance improvements for representative applications.