APB: Accelerating Distributed Long-Context Inference by Passing Compressed Context Blocks across GPUs
While long-context inference is crucial for advancing large language model (LLM) applications, its prefill speed remains a significant bottleneck. Current approaches, including sequence parallelism strategies and compute reduction through approximate attention mechanisms, still fall short of delivering optimal inference efficiency. This hinders scaling the inputs to longer sequences and processing long-context queries in a timely manner. To address this, we introduce APB, an efficient long-context inference framework that leverages multi-host approximate attention to enhance prefill speed by reducing compute and enhancing parallelism simultaneously. APB introduces a communication mechanism for essential key-value pairs within a sequence parallelism framework, enabling a faster inference speed while maintaining task performance. We implement APB by incorporating a tailored FlashAttn kernel alongside optimized distribution strategies, supporting diverse models and parallelism configurations. APB achieves speedups of up to 9.2x, 4.2x, and 1.6x compared with FlashAttn, RingAttn, and StarAttn, respectively, without any observable task performance degradation. We provide the implementation and experiment code of APB in https://github.com/thunlp/APB.
Code (1)
Tasks
Language ModelingLanguage ModellingLarge Language ModelMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
LycheeDecode: Accelerating Long-Context LLM Inference via Hybrid-Head Sparse Decoding
The proliferation of long-context large language models (LLMs) exposes a key bottleneck: the rapidly expanding key-value cache during decoding, which imposes heavy memory and latency costs. While recent approaches attemp…
Long-Context UnderstandingSlimInfer: Accelerating Long-Context LLM Inference via Dynamic Token Pruning
Long-context inference for Large Language Models (LLMs) is heavily limited by high computational demands. While several existing methods optimize attention computation, they still process the full set of hidden states at…
AdaSkip: Adaptive Sublayer Skipping for Accelerating Long-Context LLM Inference
Long-context large language models (LLMs) inference is increasingly critical, motivating a number of studies devoted to alleviating the substantial storage and computational costs in such scenarios. Layer-wise skipping m…
TawPipe: Topology-Aware Weight Pipeline Parallelism for Accelerating Long-Context Large Models Training
Training large language models (LLMs) is fundamentally constrained by limited device memory and costly inter-device communication. Although pipeline parallelism alleviates memory pressure by partitioning models across de…
Mustafar: Promoting Unstructured Sparsity for KV Cache Pruning in LLM Inference
We demonstrate that unstructured sparsity significantly improves KV cache compression for LLMs, enabling sparsity levels up to 70% without compromising accuracy or requiring fine-tuning. We conduct a systematic explorati…