DeInfer: Efficient Parallel Inferencing for Decomposed Large Language Models
For researchers and engineers scaling LLMs via decomposition, DeInfer provides a practical solution to a critical performance bottleneck.
DeInfer addresses poor parallel inference performance in decomposed large language models, achieving significant speedups through multiple optimizations while maintaining compatibility with existing techniques.
Existing works on large language model (LLM) decomposition mainly focus on improving performance on downstream tasks, but they ignore the poor parallel inference performance when trying to scale up the model size. To mitigate this important performance issue, this paper introduces DeInfer, a high-performance inference system dedicated to parallel inference of decomposed LLMs. It consists of multiple optimizations to maximize performance and be compatible with state-of-the-art optimization techniques. Extensive experiments are carried out to evaluate DeInfer's performance, where the results demonstrate its superiority, suggesting it can greatly facilitate the parallel inference of decomposed LLMs.