SLATE: Service Layer Traffic Engineering

This repository houses all the components of Service Layer Traffic Engineering (SLATE), a system that globally optimizes the flow of requests for end-to-end application latency and cost deployed in multi-region enviroments.

See the NSDI '26 Paper 🚀🚀🚀🚀🚀

See the HotNets '24 Paper 🚀🚀🚀🚀🚀

To find the multi-cluster survey results, go to multicluster survey result.pdf in the home directory of this repo.

There are three major components:

• Global controller: /global-controller
• Data plane: /wasm-plugins/slate-plugin
• Cluster controller: /cluster-controller (optional)

This system is tested on Istio.

The workflow is as such:

• The data plane in every region collects realtime metrics about current load (RPS) and experienced latency of those requests.
• The global controller builds a load to latency model for each service over time. When feasible, it uses this model in a Mixed-Integer Linear Program and solves for optimal request flow, minimizing latency. This solution is expressed as rules that are pushed to the data plane.
• The data plane enforces these rules.

Citation

@inproceedings {316074,
author = {Gangmuk Lim and Aditya Prerepa and Brighten Godfrey and Radhika Mittal},
title = {{SLATE}: Service Layer Traffic Engineering},
booktitle = {23rd USENIX Symposium on Networked Systems Design and Implementation (NSDI 26)},
year = {2026},
isbn = {978-1-939133-54-0},
address = {Renton, WA},
pages = {1895--1918},
url = {https://www.usenix.org/conference/nsdi26/presentation/lim},
publisher = {USENIX Association},
month = may
}

@inproceedings{lim2024opportunities,
  title={Opportunities and Challenges in Service Layer Traffic Engineering},
  author={Lim, Gangmuk and Prerepa, Aditya and Godfrey, Brighten and Mittal, Radhika},
  booktitle={Proceedings of the 23rd ACM Workshop on Hot Topics in Networks},
  pages={352--359},
  year={2024}
}