4 papers · 1 filter
Surviving Partial Rank Failures in Wide Expert-Parallel MoE Inference
Xun Sun, Shaoyuan Chen, Pingchuan Ma +18
Mixture-of-Experts (MoE) serving relies on wide expert parallelism (EP) to aggregate the memory capacity and bandwidth of many GPUs within one inference instance. This efficiency c…
MoFa: A Unified Performance Modeling Framework for LLM Pretraining
Lu Zhao, Rong Shi, Shaoqing Zhang +14
The exponential growth in LLM scales, with parameters soaring from billions to trillions, has necessitated distributed pretraining across large clusters comprising thousands to ten…
MemFine: Memory-Aware Fine-Grained Scheduling for MoE Training
Lu Zhao, Rong Shi, Shaoqing Zhang +21
The training of large-scale Mixture of Experts (MoE) models faces a critical memory bottleneck due to severe load imbalance caused by dynamic token routing. This imbalance leads to…
Disaggregated Prefill and Decoding Inference System for Large Language Model Serving on Multi-Vendor GPUs
Xing Chen, Rong Shi, Lu Zhao +4
LLM-based applications have been widely used in various industries, but with the increasing of models size, an efficient large language model (LLM) inference system is an urgent pr…