Related papers: Analysis and Evaluation of Using Microsecond-Laten…
The storage manager, as a key component of the database system, is responsible for organizing, reading, and delivering data to the execution engine for processing. According to the data serving mechanism, existing storage managers are…
Efficient inference of large language models (LLMs) is hindered by an ever-growing key-value (KV) cache, making KV cache compression a critical research direction. Traditional methods selectively evict less important KV cache entries, which…
The Key-Value (KV) cache in generative large language models (LLMs) introduces substantial memory overhead. Existing works mitigate this burden by offloading or compressing the KV cache. However, loading the entire cache incurs significant…
Several emerging non-volatile (NV) memory technologies are rising as interesting alternatives to build the Last-Level Cache (LLC). Their advantages, compared to SRAM memory, are higher density and lower static power, but write operations…
LLM inference for enterprise applications, such as summarization, RAG, and code-generation, typically observe much longer prompt than generations, leading to high prefill cost and response latency. We present SwiftKV, a novel model…
Shared virtual memory (SVM) is key in heterogeneous systems on chip (SoCs), which combine a general-purpose host processor with a many-core accelerator, both for programmability and to avoid data duplication. However, SVM can bring a…
Long-context LLMs have enabled numerous downstream applications but also introduced significant challenges related to computational and memory efficiency. To address these challenges, optimizations for long-context inference have been…
The linear growth of key-value (KV) cache memory and quadratic computational in attention mechanisms complexity pose significant bottlenecks for large language models (LLMs) in long-context processing. While existing KV cache optimization…
Despite the significant success of large language models (LLMs), their extensive memory requirements pose challenges for deploying them in long-context token generation. The substantial memory footprint of LLM decoders arises from the…
Large language models (LLMs) rely on key-value (KV) caches for efficient autoregressive decoding; however, cache size grows linearly with context length and model depth, becoming a major bottleneck in long-context inference. Prior KV cache…
Context lengths of Large Language Models (LLMs) have exploded in recent years, with 128k-token context becoming a standard and million-token context becoming a reality. Efficiently supporting long-context inference remains challenging as…
The computational and memory demands of large language models for generative inference present significant challenges for practical deployment. One promising solution targeting offline inference is offloading-based batched inference, which…
As its price per bit drops, SSD is increasingly becoming the default storage medium for cloud application databases. However, it has not become the preferred storage medium for key-value caches, even though SSD offers more than 10x lower…
Under modern test-time compute and agentic paradigms, language models process ever-longer sequences. Efficient text generation with transformer architectures is increasingly constrained by the Key-Value cache memory footprint and bandwidth.…
We demonstrate that unstructured sparsity significantly improves KV cache compression for LLMs, enabling sparsity levels up to 70% without compromising accuracy or requiring fine-tuning. We conduct a systematic exploration of pruning…
We present Keigo, a concurrency- and workload-aware storage middleware that enhances the performance of log-structured merge key-value stores (LSM KVS) when they are deployed on a hierarchy of storage devices. The key observation behind…
The storage stack in the traditional operating system is primarily optimized towards improving the CPU utilization and hiding the long I/O latency imposed by the slow I/O devices such as hard disk drivers (HDDs). However, the emerging…
Vision-Language Models (VLMs) have demonstrated impressive performance across a versatile set of tasks. A key challenge in accelerating VLMs is storing and accessing the large Key-Value (KV) cache that encodes long visual contexts, such as…
Large Language Models (LLMs) require significant GPU memory when processing long texts, with the key value (KV) cache consuming up to 70\% of total memory during inference. Although existing compression methods reduce memory by evaluating…
Retrieval-Augmented Generation (RAG) systems enhance the performance of large language models (LLMs) by incorporating supplementary retrieved documents, enabling more accurate and context-aware responses. However, integrating these external…