Related papers: PipeBoost: Resilient Pipelined Architecture for Fa…

Pipeline parallelism has emerged as a predominant approach for deploying large language models (LLMs) across distributed nodes, owing to its lower communication overhead compared to tensor parallelism. While demonstrating high throughput in…

Serverless computing offers a compelling cloud model for online inference services. However, existing serverless platforms lack efficient support for GPUs, hindering their ability to deliver high-performance inference. In this paper, we…

Mixture-of-Experts is a promising approach for edge AI with low-batch inference. Yet, on-device deployments often face limited on-chip memory and severe workload imbalance; the prevalent use of offloading further incurs off-chip memory…

Two widely adopted techniques for LLM inference serving systems today are hybrid batching and disaggregated serving. A hybrid batch combines prefill and decode tokens of different requests in the same batch to improve resource utilization…

The attention layer, a core component of Transformer-based LLMs, brings out inefficiencies in current GPU systems due to its low operational intensity and the substantial memory requirements of KV caches. We propose a High-bandwidth…

This paper introduces LLMServingSim2.0, a system simulator designed for exploring heterogeneous hardware in large-scale LLM serving systems. LLMServingSim2.0 addresses two key limitations of its predecessor: (1) integrating hardware models…

Modern deployment of large language models (LLMs) frequently involves both inference serving and continuous retraining to stay aligned with evolving data and user feedback. Common practices separate these workloads onto distinct servers in…

In cloud machine learning (ML) inference systems, providing low latency to end-users is of utmost importance. However, maximizing server utilization and system throughput is also crucial for ML service providers as it helps lower the…

While deploying large language models on edge devices promises low-latency and privacy-preserving AI services, it is hindered by limited device resources. Although pipeline parallelism facilitates distributed inference, existing approaches…

The transition from standard generative AI to \emph{reasoning-centric architectures}, exemplified by models capable of extensive Chain-of-Thought~(CoT) processing, marks a fundamental paradigm shift in system requirements. Unlike…

Large Language Model (LLM) inference services demand exceptionally high availability and low latency, yet multi-GPU Tensor Parallelism (TP) makes them vulnerable to single-GPU failures. We present AnchorTP, a state-preserving elastic TP…

Large language model (LLM) inference performance is increasingly bottlenecked by the memory wall. While GPUs continue to scale raw compute throughput, they struggle to deliver scalable performance for memory bandwidth bound workloads. This…

Recently, Mixture-of-Experts (MoE) has become one of the most popular techniques to scale pre-trained models to extraordinarily large sizes. Dynamic activation of experts allows for conditional computation, increasing the number of…

Large language models (LLMs) have revolutionized the field of AI, demonstrating unprecedented capacity across various tasks. However, the inference process for LLMs comes with significant computational costs. In this paper, we propose an…

The widespread adoption of LLMs has driven an exponential rise in their deployment, imposing substantial demands on inference clusters. These clusters must handle numerous concurrent queries for different LLM downstream tasks. To handle…

The growing demand for real-time processing tasks is driving the need for multi-model inference pipelines on edge devices. However, cost-effectively deploying these pipelines while optimizing Quality of Service (QoS) and costs poses…

Serverless computing has emerged as a pivotal paradigm for deploying Deep Learning (DL) models, offering automatic scaling and cost efficiency. However, the inherent cold start problem in serverless ML inference systems, particularly the…

Large language models (LLMs) power a new generation of interactive AI applications exemplified by ChatGPT. The interactive nature of these applications demands low latency for LLM inference. Existing LLM serving systems use…

Low-Rank Adaptation (LoRA) has become the leading Parameter-Efficient Fine-Tuning (PEFT) method for Large Language Models (LLMs), as it significantly reduces GPU memory usage while maintaining competitive fine-tuned model quality on…

Pipeline parallelism is widely used to train large language models (LLMs). However, increasing heterogeneity in model architectures exacerbates pipeline bubbles, thereby reducing training efficiency. Existing approaches overlook the…