Presentation
MoNDE: Mixture of Near-Data Experts for Large-Scale Sparse Models
DescriptionMixture-of-Experts (MoE) large language models (LLM) have memory requirements that often exceed the GPU memory capacity, requiring costly parameter movement from secondary memories to the GPU for expert computation. In this work, we present Mixture of Near-Data Experts (MoNDE), a near-data computing solution that efficiently enables MoE LLM inference. MoNDE reduces the volume of MoE parameter movement by transferring only the hot experts to the GPU, while computing the remaining cold experts inside the host memory device. By replacing the transfers of massive expert parameters with the ones of small activations, MoNDE enables far more communication-efficient MoE inference, thereby resulting in substantial speedups over the existing parameter offloading frameworks for both encoder and decoder operations.
Event Type
Research Manuscript
TimeWednesday, June 2611:00am - 11:15am PDT
Location3003, 3rd Floor
Design
In-memory and Near-memory Computing Architectures, Applications and Systems