Accelerating Multi-GPU Embedding Retrieval with PGAS-Style Communication for Deep Learning Recommendation Systems

In this paper, we propose using Partitioned Global Address Space (PGAS) GPU one-sided asynchronous small messages to replace the widely used collective communication calls for sparse input multi-GPU embedding retrieval in deep learning recommendation systems. This GPU PGAS communication approach ach...

Celý popis

Podrobná bibliografia

Vydané v:	SC24-W: Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis s. 1262 - 1273
Hlavní autori:	Chen, Yuxin, Buluc, Aydin, Yelick, Katherine, Owens, John D.
Médium:	Konferenčný príspevok..
Jazyk:	English
Vydavateľské údaje:	IEEE 17.11.2024
Predmet:	Analytical models Collective Calls Communication Computational modeling Conferences Deep learning Deep Learning Recommendation Model Electronics packaging Graphics processing units High performance computing PGAS Recommender systems
On-line prístup:	Získať plný text
Tagy:	Pridať tag Žiadne tagy, Buďte prvý, kto otaguje tento záznam!