Klotski: Efficient Mixture-of-Expert Inference via Expert-Aware Multi-Batch Pipeline

• URL: http://arxiv.org/abs/2502.06888v1
• Date: Sun, 09 Feb 2025 08:47:06 GMT
• Title: Klotski: Efficient Mixture-of-Expert Inference via Expert-Aware Multi-Batch Pipeline
• Authors: Zhiyuan Fang, Yuegui Huang, Zicong Hong, Yufeng Lyu, Wuhui Chen, Yue Yu, Fan Yu, Zibin Zheng,
• Abstract summary: Mixture of Experts (MoE) enables the scaling of language models up to trillions of parameters without significantly increasing computational costs. offloading techniques utilise memory from the CPU and disk and parallelise the I/O and computation for efficiency. We propose Klotski, an efficient MoE inference engine that significantly reduces pipeline bubbles through a novel expert-aware multi-batch pipeline paradigm.
• Score: 39.52960562420227
• License:
• Abstract: Mixture of Experts (MoE), with its distinctive sparse structure, enables the scaling of language models up to trillions of parameters without significantly increasing computational costs. However, the substantial parameter size presents a challenge for inference, as the expansion in GPU memory cannot keep pace with the growth in parameters. Although offloading techniques utilise memory from the CPU and disk and parallelise the I/O and computation for efficiency, the computation for each expert in MoE models is often less than the I/O, resulting in numerous bubbles in the pipeline. Therefore, we propose Klotski, an efficient MoE inference engine that significantly reduces pipeline bubbles through a novel expert-aware multi-batch pipeline paradigm. The proposed paradigm uses batch processing to extend the computation time of the current layer to overlap with the loading time of the next layer. Although this idea has been effectively applied to dense models, more batches may activate more experts in the MoE, leading to longer loading times and more bubbles. Thus, unlike traditional approaches, we balance computation and I/O time and minimise bubbles by orchestrating their inference orders based on their heterogeneous computation and I/O requirements and activation patterns under different batch numbers. Moreover, to adapt to different hardware environments and models, we design a constraint-sensitive I/O-compute planner and a correlation-aware expert prefetcher for a schedule that minimises pipeline bubbles. Experimental results demonstrate that Klotski achieves a superior throughput-latency trade-off compared to state-of-the-art techniques, with throughput improvements of up to 85.12x.

Related papers

• DSMoE: Matrix-Partitioned Experts with Dynamic Routing for Computation-Efficient Dense LLMs [70.91804882618243]
  This paper proposes DSMoE, a novel approach that achieves sparsification by partitioning pre-trained FFN layers into computational blocks. We implement adaptive expert routing using sigmoid activation and straight-through estimators, enabling tokens to flexibly access different aspects of model knowledge. Experiments on LLaMA models demonstrate that under equivalent computational constraints, DSMoE achieves superior performance compared to existing pruning and MoE approaches.
  arXiv  Detail & Related papers  (2025-02-18T02:37:26Z)
• MultiPDENet: PDE-embedded Learning with Multi-time-stepping for Accelerated Flow Simulation [48.41289705783405]
  We propose a PDE-embedded network with multiscale time stepping (MultiPDENet) In particular, we design a convolutional filter based on the structure of finite difference with a small number of parameters to optimize. A Physics Block with a 4th-order Runge-Kutta integrator at the fine time scale is established that embeds the structure of PDEs to guide the prediction.
  arXiv  Detail & Related papers  (2025-01-27T12:15:51Z)
• BATON: Enhancing Batch-wise Inference Efficiency for Large Language Models via Dynamic Re-batching [4.610983384440473]
  We propose BATON, an efficient batch-wise LLM inference scheme by dynamically adjusting processing batch. Compared to the state-of-the-art solution Orca, BATON improves query processing by up to 1.75 times.
  arXiv  Detail & Related papers  (2024-10-24T12:53:39Z)
• ExpertFlow: Optimized Expert Activation and Token Allocation for Efficient Mixture-of-Experts Inference [41.41316718220569]
  ExpertFlow is designed to enhance inference efficiency by accommodating flexible routing and enabling efficient expert scheduling between CPU and GPU. Our experiments demonstrate that ExpertFlow achieves up to 93.72% GPU memory savings and enhances inference speed by 2 to 10 times compared to baseline methods.
  arXiv  Detail & Related papers  (2024-10-23T15:24:54Z)
• EPS-MoE: Expert Pipeline Scheduler for Cost-Efficient MoE Inference [49.94169109038806]
  This paper introduces EPS-MoE, a novel expert pipeline scheduler for MoE that surpasses the existing parallelism schemes. Our results demonstrate at most 52.4% improvement in prefill throughput compared to existing parallel inference methods.
  arXiv  Detail & Related papers  (2024-10-16T05:17:49Z)
• The Missing U for Efficient Diffusion Models [3.712196074875643]
  Diffusion Probabilistic Models yield record-breaking performance in tasks such as image synthesis, video generation, and molecule design. Despite their capabilities, their efficiency, especially in the reverse process, remains a challenge due to slow convergence rates and high computational costs. We introduce an approach that leverages continuous dynamical systems to design a novel denoising network for diffusion models.
  arXiv  Detail & Related papers  (2023-10-31T00:12:14Z)
• Decreasing the Computing Time of Bayesian Optimization using Generalizable Memory Pruning [56.334116591082896]
  We show a wrapper of memory pruning and bounded optimization capable of being used with any surrogate model and acquisition function. Running BO on high-dimensional or massive data sets becomes intractable due to this time complexity. All model implementations are run on the MIT Supercloud state-of-the-art computing hardware.
  arXiv  Detail & Related papers  (2023-09-08T14:05:56Z)
• SqueezeLLM: Dense-and-Sparse Quantization [80.32162537942138]
  Main bottleneck for generative inference with LLMs is memory bandwidth, rather than compute, for single batch inference. We introduce SqueezeLLM, a post-training quantization framework that enables lossless compression to ultra-low precisions of up to 3-bit. Our framework incorporates two novel ideas: (i) sensitivity-based non-uniform quantization, which searches for the optimal bit precision assignment based on second-order information; and (ii) the Dense-and-Sparse decomposition that stores outliers and sensitive weight values in an efficient sparse format.
  arXiv  Detail & Related papers  (2023-06-13T08:57:54Z)
• DACT-BERT: Differentiable Adaptive Computation Time for an Efficient BERT Inference [3.375478015832455]
  We propose DACT-BERT, a differentiable adaptive computation time strategy for BERT-like models. DACT-BERT adds an adaptive computational mechanism to BERT's regular processing pipeline, which controls the number of Transformer blocks that need to be executed at inference time. Our experiments demonstrate that our approach, when compared to the baselines, excels on a reduced computational regime and is competitive in other less restrictive ones.
  arXiv  Detail & Related papers  (2021-09-24T04:45:55Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.