apply(recsys) Conference 2022 | Real-Time Recommendation System With Collision-less Embedding Table

apply(recsys) Conference 2022 | Monolith: Real-Time Recommendation System With Collision-less Embedding Table
by:
Youlong Cheng, Engineering Leader, ByteDance
We’ll provide an introduction to Monolith, a system tailored for online training. Our design has been driven by observations of our application workloads and production environment that reflects a marked departure from other recommendations systems. Our contributions are manifold: first, we crafted a collisionless embedding table with optimizations such as expirable embeddings and frequency filtering to reduce its memory footprint; second, we provide an production-ready online training architecture with high fault-tolerance; finally, we proved that system reliability could be traded-off for real-time learning. Monolith has successfully landed in the BytePlus Recommend product.
apply(): The ML data engineering Conference
Presented by Tecton
Connect with us:
Slack: slack.feast.dev/
LinkedIn: / tect. .

[@reemmasoud7337]

Is Monolith compatible with PyTorch also?