PRISM: Fast Online LLM Serving via Scheduling-Memory Co-design

• URL: http://arxiv.org/abs/2605.08581v1
• Date: Sat, 09 May 2026 00:48:25 GMT
• Title: PRISM: Fast Online LLM Serving via Scheduling-Memory Co-design
• Abstract summary: We present PRISM, which co-designs a query-aware scheduler (QAS) with a demand-aware rad tree (DART) to align request admission with exact- KV retention.<n>Our evaluation results show that, versus the strongest baseline, PRISM reduces average per-QPS P99 TTFT by 23.3% and 37.1% while increasing exact- KV-cache hit rate by 5.9 and 12.2 percentage points on 4B and 13B models, respectively.
• Score: 5.476189551376108
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Modern online large language model (LLM) services, such as Retrieval-Augmented Generation (RAG) and agent systems, increasingly expose two prominent characteristics: prompt segmentation (e.g., system instructions, retrieved passages, tool outputs) and hotspot skew, where a small set of these segments recurs frequently across user requests. Failing to jointly exploit these patterns could lead to repeated prefill of hot segments and prolonged TTFT, undermining both throughput and user-perceived responsiveness. However, existing work tackles these patterns independently: KV-cache management mainly exploits segment reuse while scheduling reorders requests to improve cache locality, yet neither aligns request admission with KV-cache retention. To address this gap, we first analyze how scheduling and KV-cache management jointly affect TTFT. Guided by this, we present PRISM (Prefix Reuse Optimization Integrated Scheduling and Memory), which co-designs a query-aware scheduler (QAS) with a demand-aware radix tree (DART) to align request admission with exact-prefix KV retention. Our evaluation results show that, versus the strongest baseline, PRISM reduces average per-QPS P99 TTFT by 23.3\% and 37.1\% while increasing exact-prefix KV-cache hit rate by 5.9 and 12.2 percentage points on 4B and 13B models, respectively.

Related papers

• BeaconKV: Key-Value Cache Compression Guided by Beacon Queries for Efficient Large Reasoning Model Inference [20.58039191111583]
  Large Reasoning Models (LRMs) achieve superior problem-solving through extended Chain-of-Thought (CoT) generation.<n>The resulting key-value ( KV) cache grows linearly with sequence length and creates severe memory bottlenecks.<n>We propose BeaconKV, a training-free KV cache compression method that maintains beacon queries, compact representatives for each global query cluster.
  arXiv  Detail & Related papers  (2026-09-04T10:23:16Z)
• CateKV: On Sequential Consistency for Long-Context LLM Inference Acceleration [67.17709877327232]
  Cate KV is a hybrid KV cache method that retains only critical token information for consistent heads.<n>Cate KV reduces memory usage by up to $2.72times$ and accelerates decoding by $2.18times$ in single-sample inputs.
  arXiv  Detail & Related papers  (2026-08-31T06:02:37Z)
• Kwai Summary Attention Technical Report [69.40814939510126]
  Long-context ability, has become one of the most important direction of next-generation Large Language Models.<n>Standard softmax attention exhibits quadratic time complexity with respect to sequence length.<n>We propose Kwai Summary Attention (KSA), a novel attention mechanism that reduces sequence modeling cost by compressing historical contexts.
  arXiv  Detail & Related papers  (2026-04-27T12:59:53Z)
• Stochastic KV Routing: Enabling Adaptive Depth-Wise Cache Sharing [29.913403615975174]
  Serving transformer language models with high throughput requires caching Key-Values ( KVs) to avoid redundant computation.<n>The memory footprint of KV caching is significant and heavily impacts serving costs.<n>We propose a simple training approach: random cross-layer attention. During training, layers randomly choose to attend either to their own KV states or those of a preceding layer.
  arXiv  Detail & Related papers  (2026-04-03T14:56:17Z)
• Staggered Batch Scheduling: Co-optimizing Time-to-First-Token and Throughput for High-Efficiency LLM Inference [17.27010833526918]
  Staggered Batch Scheduling (SBS) buffers requests to form optimal execution batches.<n> Load-Aware Global Allocation strategy balances computational load across DP units for both Prefill and Decode phases.<n>Our system reduces TTFT by 30%-40% and improves throughput by 15%-20% compared to state-of-the-art immediate scheduling baselines.
  arXiv  Detail & Related papers  (2025-12-18T03:45:05Z)
• CacheClip: Accelerating RAG with Effective KV Cache Reuse [8.016679032026824]
  CacheClip is a novel framework that achieves both fast TTFT and high generation quality.<n>It integrates three techniques: (1) auxiliary-model-guided token selection for selective KV cache recomputation, where the auxiliary model is finetuned to improve selection accuracy, (2) shared prefixes to eliminate redundant attention sinks, and (3) grouping strategy to maintain local coherence.
  arXiv  Detail & Related papers  (2025-10-11T09:28:26Z)
• GUI-KV: Efficient GUI Agents via KV Cache with Spatio-Temporal Awareness [75.00019285120878]
  Key-value (KV) caching can mitigate this, but storing the full cache is prohibitive for image-heavy contexts.<n>Existing cache-compression methods are sub-optimal as they do not account for the spatial and temporal redundancy of GUIs.<n>We introduce GUI-KV, a plug-and-play KV cache compression method for GUI agents that requires no retraining.
  arXiv  Detail & Related papers  (2025-10-01T05:37:54Z)
• Runtime Adaptive Pruning for LLM Inference [7.5252252615137225]
  We propose RAP, an elastic pruning framework driven by reinforcement learning (RL)<n>RAP tracks the evolving ratio between model parameters and KV-cache across practical execution.<n>RAP outperforms state-of-the-art baselines, marking the first time to jointly consider model weights and KVcache on the fly.
  arXiv  Detail & Related papers  (2025-05-22T06:12:42Z)
• QuantSpec: Self-Speculative Decoding with Hierarchical Quantized KV Cache [67.84112700032007]
  Large Language Models (LLMs) are increasingly being deployed on edge devices for long-context settings.<n>In these scenarios, the Key-Value ( KV) cache is the primary bottleneck in terms of both GPU memory and latency.<n>We propose a novel self-speculative decoding framework, QuantSpec, where the draft model shares the architecture of the target model but employs a hierarchical 4-bit quantized KV cache and 4-bit quantized weights for acceleration.
  arXiv  Detail & Related papers  (2025-02-05T20:43:48Z)
• LoRC: Low-Rank Compression for LLMs KV Cache with a Progressive Compression Strategy [59.1298692559785]
  Key-Value ( KV) cache is crucial component in serving transformer-based autoregressive large language models (LLMs) Existing approaches to mitigate this issue include: (1) efficient attention variants integrated in upcycling stages; (2) KV cache compression at test time; and (3) KV cache compression at test time. We propose a low-rank approximation of KV weight matrices, allowing plug-in integration with existing transformer-based LLMs without model retraining. Our method is designed to function without model tuning in upcycling stages or task-specific profiling in test stages.
  arXiv  Detail & Related papers  (2024-10-04T03:10:53Z)
• ThinK: Thinner Key Cache by Query-Driven Pruning [63.13363917871414]
  Large Language Models (LLMs) have revolutionized the field of natural language processing, achieving unprecedented performance across a variety of applications.<n>This paper focuses on the long-context scenario, addressing the inefficiencies in KV cache memory consumption during inference.<n>We propose ThinK, a novel query-dependent KV cache pruning method designed to minimize attention weight loss while selectively pruning the least significant channels.
  arXiv  Detail & Related papers  (2024-07-30T17:59:08Z)

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.