HPipe: Large Language Model Pipeline Parallelism for Long Context on Heterogeneous Cost-effective Devices

Abstract

Micro-enterprises and individual developers emerge long context analysis demands with powerful Large Language Models (LLMs). They try to deploy the LLMs at local, but only possess various commodity devices and the unreliable interconnection between devices. Existing parallel techniques can not fully perform in limited environment. The heterogeneity of devices, coupled with their limited capacity and expensive communication, brings challenges to private deployment for maximized utilization of available devices while masking latency. Hence, we introduce HPipe, a pipeline inference framework that successfully mitigates LLMs from high-performance clusters to heterogeneous commodity devices. By ensuring a balanced distribution of workloads, HPipe facilitates the inference through pipelining the sequences on the token dimension. The evaluation conducted on LLaMA-7B and GPT3-2B demonstrates that HPipe holds the potential for long context analysis on LLM with heterogeneity devices, achieving an impressive speedup in latency and throughput up to 2.28 times.