hipersys-team/lightning

[SIGCOMM 2023] Lightning: A Reconfigurable Photonic-Electronic SmartNIC for Fast and Energy-Efficient Inference

35
/ 100
Emerging

This project offers a unique solution for quickly processing deep neural network inference requests using a specialized SmartNIC. It takes in real-time data packets needing inference and outputs the processed results much faster and with greater energy efficiency than traditional digital systems. Network engineers and data center operators who manage high-volume, real-time AI workloads will find this particularly useful.

No commits in the last 6 months.

Use this if you need to accelerate deep neural network inference directly on network traffic with exceptional speed and energy efficiency, bypassing common bottlenecks.

Not ideal if your primary goal is general-purpose data processing or if you do not operate with real-time, high-throughput AI inference demands at the network edge.

network-acceleration real-time-AI data-center-operations deep-learning-inference smart-NIC
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 16 / 25
Community 11 / 25

How are scores calculated?

Stars

44

Forks

5

Language

Verilog

License

MIT

Last pushed

Nov 17, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/hipersys-team/lightning"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.