Related papers: Accelerating LSM-Tree with the Dentry Management o…
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…
Persistent key-value (KV) stores are critical infrastructure for data-intensive applications. Leveraging high-performance Non-Volatile Memory (NVM) to enhance KV stores has gained traction. However, previous work has primarily focused on…
In recent years, emerging storage hardware technologies have focused on divergent goals: better performance or lower cost-per-bit. Correspondingly, data systems that employ these technologies are typically optimized either to be fast (but…
Log-Structured Merge-Trees (LSM-trees) dominate persistent key-value storage but suffer from high write amplification from 10x to 30x under random workloads due to repeated compaction. This overhead becomes prohibitive for large values with…
In this paper we present a new approach of incorporating kernels into dictionary learning. The kernel K-SVD algorithm (KKSVD), which has been introduced recently, shows an improvement in classification performance, with relation to its…
Long Short-Term Memory (LSTM) networks, a type of recurrent neural network with a more complex computational unit, have been successfully applied to a variety of sequence modeling tasks. In this paper we develop Tree Long Short-Term Memory…
Vision-Language Models (VLMs) are integral to tasks such as image captioning and visual question answering, but their high computational cost, driven by large memory footprints and processing time, limits their scalability and real-time…
Log-Structured Merge (LSM) tree-based Key-Value Stores (KVSs) are widely adopted for their high performance in write-intensive environments, but they often face performance degradation due to write stalls during compaction. Prior solutions,…
We introduce the concept of design continuums for the data layout of key-value stores. A design continuum unifies major distinct data structure designs under the same model. The critical insight and potential long-term impact is that such…
State-of-the-art Large Language Models (LLMs) excel in code generation at the function level. However, the output quality significantly declines when scaling to repository-level systems. Current workflows relying only on natural language…
LRM-Trees are an elegant way to partition a sequence of values into sorted consecutive blocks, and to express the relative position of the first element of each block within a previous block. They were used to encode ordinal trees and to…
Although an increasing number of databases now embrace shared-storage architectures, current storage-disaggregated systems have yet to strike an optimal balance between cost and performance. In high-concurrency read/write scenarios,…
Data processing engines increasingly leverage distributed file systems for scalable, cost-effective storage. While the Apache Parquet columnar format has become a popular choice for data storage and retrieval, the immutability of Parquet…
We consider multi-label classification where the goal is to annotate each data point with the most relevant $\textit{subset}$ of labels from an extremely large label set. Efficient annotation can be achieved with balanced tree predictors,…
Containerization simplifies the sharing and deployment of applications when environments change in the software delivery chain. To deploy an application, container delivery methods push and pull container images. These methods operate on…
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…
Recent advances in long-text understanding have pushed the context length of large language models (LLMs) up to one million tokens. It boosts LLMs's accuracy and reasoning capacity but causes exorbitant computational costs and…
Recent large vision-language models (LVLMs) demonstrate remarkable capabilities in processing extended multi-modal sequences, yet the resulting key-value (KV) cache expansion creates a critical memory bottleneck that fundamentally limits…
The matrix quantization entails representing matrix elements in a more space-efficient form to reduce storage usage, with dequantization restoring the original matrix for use. We formulate the Quantization Error Minimization (QEM) problem…
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…