Related papers: FOCUS: Boosting Schema-aware Access for KV Stores …

Supporting long-context LLMs is challenging due to the substantial memory demands of the key-value (KV) cache. Existing offloading systems store the full cache in host memory and selectively fetch critical entries during decoding, but this…

LSM-tree based key-value (KV) stores organize data in a multi-level structure for high-speed writes. Range queries on traditional LSM-trees must seek and sort-merge data from multiple table files on the fly, which is expensive and often…

We introduce KV-Fold, a simple, training-free long-context inference protocol that treats the key-value (KV) cache as the accumulator in a left fold over sequence chunks. At each step, the model processes the next chunk conditioned on the…

Multi-modal Large Language Models (MLLMs) serving systems commonly employ KV-cache compression to reduce memory footprint. However, existing compression methods introduce significant processing overhead and queuing delays, particularly in…

Cache fusion accelerates generation process of LLMs equipped with RAG through KV caching and selective token recomputation, thereby reducing computational costs and improving efficiency. However, existing methods primarily rely on local…

Autonomous vehicles (AVs) are evolving into mobile computing platforms, equipped with powerful processors and diverse sensors that generate massive heterogeneous data, for example 14 TB per day. Supporting emerging third-party applications…

The Key-Value (KV) cache is a crucial component in serving transformer-based autoregressive large language models (LLMs), enabling faster inference by storing previously computed KV vectors. However, its memory consumption scales linearly…

We present LearnedKV, a novel tiered key-value store that seamlessly integrates a Log-Structured Merge (LSM) tree with a Learned Index to achieve superior read and write performance on storage systems. While existing approaches use learned…

Efficiently serving large language models (LLMs) requires batching of many requests to reduce the cost per request. Yet, with larger batch sizes and longer context lengths, the key-value (KV) cache, which stores attention keys and values to…

Systems that require high-throughput and fault tolerance, such as key-value stores and databases, are looking to persistent memory to combine the performance of in-memory systems with the data-consistent fault-tolerance of nonvolatile…

Transformer decoders have achieved strong results across tasks, but the memory required for the KV cache becomes prohibitive at long sequence lengths. Although Cross-layer KV Cache sharing (e.g., YOCO, CLA) offers a path to mitigate KV…

Large language model (LLM) based agentic workflows have become a popular paradigm for coordinating multiple specialized agents to solve complex tasks. To improve serving efficiency, existing LLM systems employ prefix caching to reuse…

Flat combining (FC) is a synchronization paradigm in which a single thread, holding a global lock, collects requests by multiple threads for accessing a concurrent data structure and applies their combined requests to it. Although FC is…

Multi-tenancy hosting of users in cloud NoSQL data stores is favored by cloud providers because it enables resource sharing at low operating cost. Multi-tenancy takes several forms depending on whether the back-end file system is a local…

Key-value (KV) caching has emerged as a crucial optimization technique for accelerating inference in large language models (LLMs). By allowing the attention operation to scale linearly rather than quadratically with the total sequence…

Training effective Text-to-SQL models remains challenging due to the scarcity of high-quality, diverse, and structurally complex datasets. Existing methods either rely on limited human-annotated corpora, or synthesize datasets directly by…

Large Language Models (LLMs) excel across a variety of language tasks yet are constrained by limited input lengths and high computational costs. Existing approaches\textemdash such as relative positional encodings (e.g., RoPE, ALiBi) and…

We present our framework DKVF that enables one to quickly prototype and evaluate new protocols for key-value stores and compare them with existing protocols based on selected benchmarks. Due to limitations of CAP theorem, new protocols must…

Traditional KV cache eviction strategies, which discard less critical KV-pairs based on attention scores, often degrade generation quality, causing context loss or hallucinations. Recent efforts shift toward KV merging, merging eviction…

Key-Value (KV) cache quantization has become a widely adopted optimization technique for efficient large language models (LLMs) inference by reducing KV cache memory usage and mitigating memory-bound constraints. Recent studies have…