Related papers: VESPA: VIPT Enhancements for Superpage Accesses

Despite advances in large language model (LLM)-based natural language interfaces for databases, scaling to enterprise-level data catalogs remains an under-explored challenge. Prior works addressing this challenge rely on domain-specific…

High load latency that results from deep cache hierarchies and relatively slow main memory is an important limiter of single-thread performance. Data prefetch helps reduce this latency by fetching data up the hierarchy before it is…

Vision Transformers (ViTs) have demonstrated remarkable potential in image processing tasks by utilizing self-attention mechanisms to capture global relationships within data. However, their scalability is hindered by significant…

Vector searches on large-scale datasets are critical to modern online services like web search and RAG, which necessity storing the datasets and their index on the secondary storage like SSD. In this paper, we are the first to characterize…

Vision transformers have shown great success on numerous computer vision tasks. However, its central component, softmax attention, prohibits vision transformers from scaling up to high-resolution images, due to both the computational…

Large-scale applications of Visual Place Recognition (VPR) require computationally efficient approaches. Further, a well-balanced combination of data-based and training-free approaches can decrease the required amount of training data and…

The Locator/ID Separation Protocol (LISP) limits the growth of the Default-Free Zone routing tables by creating a highly aggregatable and quasi-static Internet core. However, LISP pushes the forwarding state to edge routers whose timely…

Hash tables are ubiquitous. Properties such as an amortized constant time complexity for insertion and querying as well as a compact memory layout make them versatile associative data structures with manifold applications. The rapidly…

In recent years, Transformer has achieved good results in Natural Language Processing (NLP) and has also started to expand into Computer Vision (CV). Excellent models such as the Vision Transformer and Swin Transformer have emerged. At the…

Modern data-driven applications expose limitations of von Neumann architectures - extensive data movement, low throughput, and poor energy efficiency. Accelerators improve performance but lack flexibility and require data transfers.…

This paper proposes a fully decentralized peer-to-peer overlay structure GeoP2P, to facilitate geographic location based search and retrieval of information. Certain limitations of centralized geographic indexes favor peer-to-peer…

The 3.2 giga-pixel LSST camera will produce approximately half a petabyte of archive images every month. These data need to be reduced in under a minute to produce real-time transient alerts, and then added to the cumulative catalog for…

Modern database clusters entail two levels of networks: connecting CPUs and NUMA regions inside a single server in the small and multiple servers in the large. The huge performance gap between these two types of networks used to slow down…

High-Performance Computing (HPC) applications need to checkpoint massive amounts of data at scale. Multi-level asynchronous checkpoint runtimes like VELOC (Very Low Overhead Checkpoint Strategy) are gaining popularity among application…

By their very name caches are often overlooked and yet play a vital role in the performance of modern and indeed future hardware. Using MAGPIE (Machine Automated General Performance Improvement via Evolution of software) we show genetic…

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…

Documents are 2-dimensional carriers of written communication, and as such their interpretation requires a multi-modal approach where textual and visual information are efficiently combined. Document Visual Question Answering (Document…

Multi-Head Latent Attention (MLA), introduced in DeepSeek-V2, compresses key-value states into a low-rank latent vector, caching only this vector to reduce memory. In tensor parallelism (TP), however, attention heads are computed across…

Over the last decade, most of the increase in computing power has been gained by advances in accelerated many-core architectures, mainly in the form of GPGPUs. While accelerators achieve phenomenal performances in various computing tasks,…

With the advancements in long-context inference capabilities of large language models (LLMs), the KV cache has become one of the foundational components. However, its substantial GPU memory consumption makes KV cache compression a key…