The Effect of Scheduling and Preemption on the Efficiency of LLM Inference Serving

• URL: http://arxiv.org/abs/2411.07447v2
• Date: Tue, 19 Nov 2024 21:57:16 GMT
• Title: The Effect of Scheduling and Preemption on the Efficiency of LLM Inference Serving
• Authors: Kyoungmin Kim, Kijae Hong, Caglar Gulcehre, Anastasia Ailamaki,
• Abstract summary: INFERMAX is an analytical framework that uses inference cost models to compare various schedulers. Our findings indicate that preempting requests can reduce GPU costs by 30% compared to avoiding preemptions at all.
• Score: 8.552242818726347
• License:
• Abstract: The growing usage of Large Language Models (LLMs) highlights the demands and challenges in scalable LLM inference systems, affecting deployment and development processes. On the deployment side, there is a lack of comprehensive analysis on the conditions under which a particular scheduler performs better or worse, with performance varying substantially across different schedulers, hardware, models, and workloads. Manually testing each configuration on GPUs can be prohibitively expensive. On the development side, unpredictable performance and unknown upper limits can lead to inconclusive trial-and-error processes, consuming resources on ideas that end up ineffective. To address these challenges, we introduce INFERMAX, an analytical framework that uses inference cost models to compare various schedulers, including an optimal scheduler formulated as a constraint satisfaction problem (CSP) to establish an upper bound on performance. Our framework offers in-depth analysis and raises essential questions, challenging assumptions and exploring opportunities for more efficient scheduling. Notably, our findings indicate that preempting requests can reduce GPU costs by 30% compared to avoiding preemptions at all. We believe our methods and insights will facilitate the cost-effective deployment and development of scalable, efficient inference systems and pave the way for cost-based scheduling.

Related papers

• Inference-Time Computations for LLM Reasoning and Planning: A Benchmark and Insights [49.42133807824413]
  We examine the reasoning and planning capabilities of large language models (LLMs) in solving complex tasks. Recent advances in inference-time techniques demonstrate the potential to enhance LLM reasoning without additional training. OpenAI's o1 model shows promising performance through its novel use of multi-step reasoning and verification.
  arXiv  Detail & Related papers  (2025-02-18T04:11:29Z)
• 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)
• PEA: Enhancing LLM Performance on Computational-Reasoning Tasks [21.13926189404758]
  This study introduces a formal approach to describe and solve a class of important reasoning tasks termed computational reasoning problems. The framework decomposes these problems into predicate and enumeration components, using LLMs to synthesize programs based on specified predicates, enumeration, and aggregation rules. Empirical evaluation reveals that PEA substantially enhances the performance of underlying models on benchmark computational problems, yielding an average accuracy improvement of approximately $50%$, coupled with increased efficiency.
  arXiv  Detail & Related papers  (2025-02-16T00:27:05Z)
• CITER: Collaborative Inference for Efficient Large Language Model Decoding with Token-Level Routing [56.98081258047281]
  CITER enables efficient collaboration between small and large language models (SLMs & LLMs) through a token-level routing strategy. We formulate router training as a policy optimization, where the router receives rewards based on both the quality of predictions and the inference costs of generation. Our experiments show that CITER reduces the inference costs while preserving high-quality generation, offering a promising solution for real-time and resource-constrained applications.
  arXiv  Detail & Related papers  (2025-02-04T03:36:44Z)
• Efficiently Serving LLM Reasoning Programs with Certaindex [4.681117143870077]
  Dynasor is a system that optimize inference-time compute for large language models (LLMs) reasoning queries. Unlike traditional engines, Dynasor tracks and schedules requests within reasoning queries. It reduces compute by up to 50% in batch processing and sustaining 3.3x higher query rates or 4.7x tighter latency SLOs in online serving.
  arXiv  Detail & Related papers  (2024-12-30T14:57:53Z)
• Leveraging Graph-RAG and Prompt Engineering to Enhance LLM-Based Automated Requirement Traceability and Compliance Checks [8.354305051472735]
  This study demonstrates that integrating a robust Graph-RAG framework with advanced prompt engineering techniques, such as Chain of Thought and Tree of Thought, can significantly enhance performance. It is both costly and more complex to implement across diverse contexts, requiring careful adaptation to specific scenarios.
  arXiv  Detail & Related papers  (2024-12-11T18:11:39Z)
• GUIDE: A Global Unified Inference Engine for Deploying Large Language Models in Heterogeneous Environments [1.0558515062670693]
  Large language models (LLMs) in real-world scenarios remains a critical challenge. These challenges often lead to inefficiencies in memory utilization, latency, and throughput. We develop a framework to address these issues, achieving prediction errors between 9.9% and 42.3% for key metrics such as batch latency, TTFT, and decode throughput.
  arXiv  Detail & Related papers  (2024-12-06T05:46:43Z)
• Unlocking Large Language Model's Planning Capabilities with Maximum Diversity Fine-tuning [10.704716790096498]
  Large language models (LLMs) have demonstrated impressive task-solving capabilities, achieved through either prompting techniques or system designs. This paper investigates the impact of fine-tuning on LLMs' planning capabilities. We propose the Maximum Diversity Fine-Tuning (MDFT) strategy to improve the sample efficiency of fine-tuning in the planning domain.
  arXiv  Detail & Related papers  (2024-06-15T03:06:14Z)
• Characterization of Large Language Model Development in the Datacenter [55.9909258342639]
  Large Language Models (LLMs) have presented impressive performance across several transformative tasks. However, it is non-trivial to efficiently utilize large-scale cluster resources to develop LLMs. We present an in-depth characterization study of a six-month LLM development workload trace collected from our GPU datacenter Acme.
  arXiv  Detail & Related papers  (2024-03-12T13:31:14Z)
• Learning Logic Specifications for Policy Guidance in POMDPs: an Inductive Logic Programming Approach [57.788675205519986]
  We learn high-quality traces from POMDP executions generated by any solver. We exploit data- and time-efficient Indu Logic Programming (ILP) to generate interpretable belief-based policy specifications. We show that learneds expressed in Answer Set Programming (ASP) yield performance superior to neural networks and similar to optimal handcrafted task-specifics within lower computational time.
  arXiv  Detail & Related papers  (2024-02-29T15:36:01Z)
• Uncertainty-aware Remaining Useful Life predictor [57.74855412811814]
  Remaining Useful Life (RUL) estimation is the problem of inferring how long a certain industrial asset can be expected to operate. In this work, we consider Deep Gaussian Processes (DGPs) as possible solutions to the aforementioned limitations. The performance of the algorithms is evaluated on the N-CMAPSS dataset from NASA for aircraft engines.
  arXiv  Detail & Related papers  (2021-04-08T08:50:44Z)

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.