ServicenavigationHauptnavigationTrailKarteikarten

Research unit
INNOSUISSE
Project number
129.586 INNO-ICT
Project title
Accelerating ML via Computational Storage

Texts for this project
 GermanFrenchItalianEnglish
Key words
-
-
-
Anzeigen
Abstract
-
-
-
Anzeigen

Inserted texts


CategoryText
Key words
(English)
ICT, Machine learning
Abstract
(English)
GPUs excel in ML but have limited HBM memory, restricting model sizes. Our devices enhance GPU memory by swiftly paging model weights, freeing up HBM for larger models or extended context in LLMs. We aim to boost our prototype's bandwidth by experimenting with faster and larger memories.