CAFE: Towards Compact, Adaptive, and Fast Embedding for Large-scale Recommendation Models
Recently, the growing memory demands of embedding tables in Deep Learning
Recommendation Models (DLRMs) pose great challenges for model training and …
Recommendation Models (DLRMs) pose great challenges for model training and …
Embedding Optimization for Training Large-scale Deep Learning Recommendation Systems with EMBark
S Liu, N Zheng, H Kang, X Simmons, J Zhang… - Proceedings of the 18th …, 2024 - dl.acm.org
Training large-scale deep learning recommendation models (DLRMs) with embedding
tables stretching across multiple GPUs in a cluster presents a unique challenge, demanding …
tables stretching across multiple GPUs in a cluster presents a unique challenge, demanding …
[PDF][PDF] DLRover-RM: Resource Optimization for Deep Recommendation Models Training in the Cloud
Deep learning recommendation models (DLRM) rely on large embedding tables to manage
categorical sparse features. Expanding such embedding tables can significantly enhance …
categorical sparse features. Expanding such embedding tables can significantly enhance …
Mixed-Precision Embeddings for Large-Scale Recommendation Models
Embedding techniques have become essential components of large databases in the deep
learning era. By encoding discrete entities, such as words, items, or graph nodes, into …
learning era. By encoding discrete entities, such as words, items, or graph nodes, into …
Disaggregating Embedding Recommendation Systems with FlexEMR
Efficiently serving embedding-based recommendation (EMR) models remains a significant
challenge due to their increasingly large memory requirements. Today's practice splits the …
challenge due to their increasingly large memory requirements. Today's practice splits the …
Fine-Grained Embedding Dimension Optimization During Training for Recommender Systems
Huge embedding tables in modern Deep Learning Recommender Models (DLRM) require
prohibitively large memory during training and inference. Aiming to reduce the memory …
prohibitively large memory during training and inference. Aiming to reduce the memory …
Towards a Flexible and High-Fidelity Approach to Distributed DNN Training Emulation
B Liu, MA Ojewale, Y Ding, M Canini - … of the 15th ACM SIGOPS Asia …, 2024 - dl.acm.org
We propose NeuronaBox, a flexible, user-friendly, and high-fidelity approach to emulate
DNN training workloads. We argue that to accurately observe performance, it is possible to …
DNN training workloads. We argue that to accurately observe performance, it is possible to …
Pushing the Performance Envelope of DNN-based Recommendation Systems Inference on GPUs
Personalized recommendation is a ubiquitous application on the internet, with many
industries and hyperscalers extensively leveraging Deep Learning Recommendation …
industries and hyperscalers extensively leveraging Deep Learning Recommendation …
ERCache: An Efficient and Reliable Caching Framework for Large-Scale User Representations in Meta's Ads System
F Zhou, Y Huang, D Liang, D Li, Z Zhang… - arXiv preprint arXiv …, 2024 - arxiv.org
The increasing complexity of deep learning models used for calculating user
representations presents significant challenges, particularly with limited computational …
representations presents significant challenges, particularly with limited computational …
Proof-of-Concept of a Flexible and High-Fidelity Approach to Distributed DNN Training Emulation
B Liu, MA Ojewale, Y Ding, M Canini - Proceedings of the 2024 …, 2024 - dl.acm.org
We propose NeuronaBox, a flexible, user-friendly, and high-fidelity approach to emulate
DNN training workloads. We argue that to accurately observe performance, it is possible to …
DNN training workloads. We argue that to accurately observe performance, it is possible to …