Llumnix：面向大语言模型服务的动态调度
Llumnix: Dynamic Scheduling for Large Language Model Serving
AI总结：
Llumnix通过跨模型实例的运行时请求重调度和实时迁移机制，解决LLM服务中请求异构不可预测导致的延迟与SLO问题，显著改善尾部延迟、加速高优先级请求并降低成本。
摘要：
大语言模型（LLMs）的推理服务是释放其在人们日常生活中潜力的关键。然而，高效的LLM服务至今仍具挑战性，因为请求在资源和延迟需求方面本质上是异构且不可预测的，这源于应用的多样性以及LLM的动态执行特性。现有系统在处理这些特性方面存在根本性局限，并导致严重的排队延迟、糟糕的尾部延迟以及服务水平目标（SLO）违约等问题。我们提出了Llumnix，这是一个LLM服务系统，通过跨多个模型实例的运行时重新调度来应对此类异构且不可预测的请求。类似于现代操作系统中跨CPU核心的上下文切换，Llumnix重新调度请求以改善负载均衡和隔离性，缓解资源碎片化，并区分请求优先级和SLO。Llumnix通过一种高效且可扩展的请求及其内存状态的实时迁移机制来实现重新调度，并将其用于一个动态调度策略中，该策略优雅地统一了多种重新调度场景。我们的评估表明，与最先进的LLM服务系统相比，Llumnix将尾部延迟提升了一个数量级，将高优先级请求加速了最高1.5倍，并在实现相似尾部延迟的同时节省了最高36%的成本。Llumnix已在https://github.com/AlibabaPAI/llumnix公开提供。
英文摘要：
Inference serving for large language models (LLMs) is the key to unleashing their potential in people's daily lives. However, efficient LLM serving remains challenging today because the requests are inherently heterogeneous and unpredictable in terms of resource and latency requirements, as a result of the diverse applications and the dynamic execution nature of LLMs. Existing systems are fundamentally limited in handling these characteristics and cause problems such as severe queuing delays, poor tail latencies, and SLO violations. We introduce Llumnix, an LLM serving system that reacts to such heterogeneous and unpredictable requests by runtime rescheduling across multiple model instances. Similar to context switching across CPU cores in modern operating systems, Llumnix reschedules requests to improve load balancing and isolation, mitigate resource fragmentation, and differentiate request priorities and SLOs. Llumnix implements the rescheduling with an efficient and scalable live migration mechanism for requests and their in-memory states, and exploits it in a dynamic scheduling policy that unifies the multiple rescheduling scenarios elegantly. Our evaluations show that Llumnix improves tail latencies by an order of magnitude, accelerates high-priority requests by up to 1.5x, and delivers up to 36% cost savings while achieving similar tail latencies, compared against state-of-the-art LLM serving systems. Llumnix is publicly available at https://github.com/AlibabaPAI/llumnix.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
