Large language model (LLM) serving is becoming an increasingly important workload for cloud providers. Based on performance SLO requirements, LLM inference requests can be divided into (a) interactive requests that have tight SLOs in the order of seconds, and (b) batch requests that have relaxed SLO in the order of minutes to hours. These SLOs can degrade based on the arrival rates, multiplexing, and configuration parameters, thus necessitating the use of resource autoscaling on serving instances and their batch sizes. However, previous autoscalers for LLM serving do not consider request SLOs leading to unnecessary scaling and resource under-utilization. To address these limitations, we introduce Chiron, an autoscaler that uses the idea of hierarchical backpressure estimated using queue size, utilization, and SLOs. Our experiments show that Chiron achieves up to 90% higher SLO attainment and improves GPU efficiency by up to 70% compared to existing solutions.

The stateful nature of large language model (LLM) servingcan easily throttle precious GPU memory under load burstor long-generation requests like chain-of-thought reasoning,causing latency spikes due to queuing incoming requests. However, state-of-the-art KVCache centric approaches handleload spikes by dropping, migrating, or swapping KVCache,which faces an essential tradeoff between the performance ofongoing vs. incoming requests and thus still severely violatesSLO.This paper makes a key observation such that model param-eters are independent of the requests and are replicated acrossGPUs, and thus proposes a parameter-centric approach byselectively dropping replicated parameters to leave preciousmemory for requests. However, LLM requires KVCache tobe saved in bound with model parameters and thus droppingparameters can cause either huge computation waste or longnetwork delay, affecting all ongoing requests. Based on the ob-servation that attention operators can be decoupled from otheroperators, this paper further proposes a novel remote attentionmechanism through pipeline parallelism so as to serve up-coming requests with the additional memory borrowed fromparameters on remote GPUs. This paper further addresses sev-eral other challenges including lively exchanging KVCachewith incomplete parameters, generating an appropriate planthat balances memory requirements with cooperative exe-cution overhead, and seamlessly restoring parameters whenthe throttling has gone. Evaluations show thatKUNSERVEreduces the tail TTFT of requests under throttling by up to 27.3x compared to the state-of-the-art.

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.