搜索结果：Redis

Hybrid quantum--classical workflows often execute large ensembles of circuits that differ syntactically but implement identical operations, leading to substantial redundant computation. To address this, we introduce the Quantum Circuit Cache, a content-addressable system that detects semantic equivalence and reuses previously computed results across executions, backends, and workflow stages. Our approach combines ZX-calculus reduction with isomorphism-invariant Weisfeiler--Leman graph hashing to generate deterministic circuit identifiers, enabling constant-time lookup in distributed caches supporting both lightweight LMDB and scalable Redis deployments. The system integrates transparently into hybrid HPC workflows and remains backend-agnostic across CPU, GPU, and QPU environments. We evaluate the system on MareNostrum 5 with two representative workloads: distributed wire cutting and Differential Evolution-based QAOA optimization. For wire cutting, caching eliminates up to 91.98% of redundant subcircuit simulations, yielding speedups up to 7.0 times on a single node and maintaining advantages at scale, with Redis-based caching achieving up to 1.6 times speedups under high parallelis

Language models rely on semantic priors to perform in-context learning, which leads to poor performance on tasks involving inductive reasoning. Instruction-tuning methods based on imitation learning can superficially enhance the in-context learning performance of language models, but they often fail to improve the model's understanding of the underlying rules that connect inputs and outputs in few-shot demonstrations. We propose ReDis, a reasoning distillation technique designed to improve the inductive reasoning capabilities of language models. Through a careful combination of data augmentation, filtering, supervised fine-tuning, and alignment, ReDis achieves significant performance improvements across a diverse range of tasks, including 1D-ARC, List Function, ACRE, and MiniSCAN. Experiments on three language model backbones show that ReDis outperforms equivalent few-shot prompting baselines across all tasks and even surpasses the teacher model, GPT-4o, in some cases. ReDis, based on the LLaMA-3 backbone, achieves relative improvements of 23.2%, 2.8%, and 66.6% over GPT-4o on 1D-ARC, ACRE, and MiniSCAN, respectively, within a similar hypothesis search space. The code, dataset, and

We present SQL Query Engine, an open-source, self-hosted service that translates natural language questions into validated PostgreSQL queries through a two-stage LLM pipeline. The first stage performs automatic schema introspection and SQL generation; a multi-strategy response parser extracts SQL from any LLM output format (JSON, code blocks, or raw text) without requiring structured output APIs. The second stage executes the query against PostgreSQL and, upon failure or empty results, enters an iterative self-healing loop in which the LLM diagnoses the error using full SQLSTATE codes and PostgreSQL diagnostic messages. Two mechanisms prevent regressions: early-accept returns successful queries immediately without LLM re-evaluation, and best-result tracking preserves the best partial result across retries. Schema context is cached per session in Redis, progress events stream via Redis Pub/Sub and SSE, and an OpenAI-compatible /v1/chat/completions endpoint lets existing tools work without modification. All database connections are read-only at the driver level. We evaluate across five LLM backends on a synthetic benchmark (75 questions, three databases) where the self-healing loop y

Modern logistics systems tend to generate continuous streams of data from sources such as GPS, IoT sensors, and logistics management systems. The aggregation, processing, and analysis of data have become vital for monitoring operations, optimizing efficiency, and responding quickly to decision making tasks. In this paper, an event-driven MapReduce framework for real-time data processing in logistics environments is presented. This system runs on Kubernetes with Knative and utilizes Apache Kafka as the backbone for communication between the components. This platform is composed of five loosely coupled services that receive, process, and aggregate the incoming data in real-time. Redis is used to preserve workflow metadata, while an AWS S3 service provides persistent storage for the framework. The design is inspired by the MapReduce programming model. It integrates Function-as-a-Service (FaaS) principles with distributed processing techniques that allow configurable scaling based on the workload demands and the underlying hardware. Experimental evaluation shows that the system can scale effectively as the input data volume increases while supporting scale-to-zero, on-demand processing

Detecting anomalies in large-scale system logs is critical for the reliability and security of modern computing infrastructure. We present LogNEO, a log anomaly detector built on EleutherAI's GPT-Neo (1.3B parameters) and fine-tuned with a novel partial-credit, exponentially decaying position-aware reward scheme combined with cross-entropy regularisation via Proximal Policy Optimisation (PPO). The position-aware reward explicitly models prediction difficulty: early positions receive higher rewards for correct predictions, while later positions incur stronger penalties for errors. LogNEO attains F1-scores of 0.927, 0.913, and 0.984 on the HDFS, BGL, and Thunderbird benchmarks, improving recall by up to 6 percentage points over the prior state-of-the-art LogGPT while maintaining comparable precision. A production microservice deployment over Apache Kafka, Redis, and TensorRT-accelerated inference demonstrates 45 ms end-to-end latency at 15,000 events per second.

Designing a rate limiter that is simultaneously accurate, available, and scalable presents a fundamental challenge in distributed systems, primarily due to the trade-offs between algorithmic precision, availability, consistency, and partition tolerance. This article presents a concrete architecture for a distributed rate limiting system in a production-grade environment. Our design chooses the in-memory cache database, the Redis, along with its Sorted Set data structure, which provides $O(log (N))$ time complexity operation for the key-value pair dataset with efficiency and low latency, and maintains precision. The core contribution is quantifying the accuracy and memory cost trade-off of the chosen Rolling Window as the implemented rate limiting algorithm against the Token Bucket and Fixed Window algorithms. In addition, we explain how server-side Lua scripting is critical to bundling cleanup, counting, and insertion into a single atomic operation, thereby eliminating race conditions in concurrent environments. In the system architecture, we propose a three-layer architecture that manages the storage and updating of the limit rules. Through script load by hashing the rule paramete

Most token reduction methods for Vision Transformers seek favorable tradeoffs between accuracy and efficiency by pruning, merging, or pooling patch tokens. REDI (Relevance for DINOv3 Token Reduction) studies this question through a controlled supervised reference: how should a fixed token budget be allocated across patches for image classification? REDI quantizes final block DINOv3 patch representations into a visual vocabulary and derives class conditioned corpus scores using supervised TF-IDF over visual words. For each validation image, the ground truth class selects a row of the TF-IDF table, and four transformed views produce a TF-IDF map aligned to a reference center crop. A separate dense pass on the same crop provides an attention map. After independent min max normalization, their elementwise product defines the REDI score. A fixed keep, merge, and compress operator then uses score rank to assign patch roles and score magnitude to weight merging and compression. With precomputed REDI scores, a frozen DINOv3 ViT-B/16 backbone, and the same linear classifier used for dense evaluation, the operator reduces the sequence length from 201 to 107 tokens, a 46.8% sequence reduction

Leadership computing facilities steward large-scale scientific datasets that routinely require substantial transformation before serving as AI training data. However, no existing framework fully unifies automated transformation, readiness assessment, provenance tracking, and agent-native deployment. We present REDI, an open-source framework that addresses this gap through a unified five-stage pipeline (ingest, preprocess, transform, structure, and output) with per-stage instrumentation for reproducibility and deployment as an agent-callable skill; companion tool SetGo automates FAIR compliance and catalog publication. Evaluated across climate, proteomics, materials science, and nuclear fusion, REDI transforms all datasets from raw to AI-ready, with outputs validated against domain-expert references, and preliminary results show near-ideal parallel scaling to 100 nodes on Frontier for the climate case. Provenance-instrumented profiling reveals file I/O as the dominant pipeline cost, with format selection a first-order optimization lever. These results establish REDI as a cross-domain platform providing automated data readiness for scientific AI, transforming data preparation bottlen

Vision Foundation Models (VFMs) have significantly advanced dense feature matching, yet severe in-plane rotation remains a critical challenge. Existing solutions face a fundamental dilemma: data-driven methods require inefficient parameter scaling to implicitly learn rotations, whereas strictly equivariant networks lack the semantic capacity of modern VFMs. Consequently, current frameworks typically freeze VFMs and shift the entire burden of rotation generalization to the downstream decoder. To break this architectural bottleneck, we propose REDI-Match, an efficient framework driven by a novel Rotation-Equivariant Distillation (REDI) paradigm. Instead of relying on rotation data augmentation to establish rotational correspondences, REDI distills the non-equivariant semantic representations of a VFM into a lightweight, strictly rotation-equivariant encoder, leveraging an equivariant geometric architecture to constrain robust high-dimensional semantics. To fully exploit these features, we equip the decoder with an entropy-driven spatial alignment module. By evaluating discrete rotation hypotheses, this mechanism explicitly locks onto the canonical coordinate system, eliminating globa

The increasing adoption of large language models (LLMs) with extended context windows necessitates efficient Key-Value Cache (KVC) management to optimize inference performance. Inference workloads like Retrieval-Augmented Generation (RAG) and agents exhibit high cache reusability, making efficient caching critical to reducing redundancy and improving speed. We analyze real-world KVC access patterns using publicly available traces and evaluate commercial key-value stores like Redis and state-of-the-art RDMA-based systems (CHIME [1] and Sherman [2]) for KVC metadata management. Our work demonstrates the lack of tailored storage solution for KVC prefilling, underscores the need for an efficient distributed caching system with optimized metadata management for LLM workloads, and provides insights into designing improved KVC management systems for scalable, low-latency inference.

This study explores the use of deep reinforcement learning (DRL) for active flow control (AFC) to reduce flow separation on wings at high angles of attack. Concretely, here the DRL agent controls the flow over the three-dimensional NACA0012 wing section at the Reynolds number Re = 1,000 and angle of attack AoA = 20 degrees, autonomously identifying optimal control actions through real-time flow data and a reward function focused on improving aerodynamic performance. The framework integrates the GPU-accelerated computational fluid dynamics (CFD) solver SOD2D with the TF-Agents DRL library via a Redis in-memory database, enabling rapid training. This work builds on previous DRL flow-control studies, demonstrating DRL potential to address complex aerodynamic challenges and push the boundaries of traditional AFC methods.

We developed the SMA eXchange (SMA-X) as a real-time data sharing solution, built atop a central Redis database. SMA-X is a storage convention, facilitated by a set of server-side Lua scripts (or Redis functions) which enable efficient low-latency and high-throughput real-time sharing of hierarchically structured data among the various systems and subsystems of the telescope. It enables fast, atomic retrievals of specific leaf elements, branches, and sub-trees, including associated metadata (types, dimensions, timestamps, and origins, and more). At the Submillimer Array (SMA) we rely on it since 2021 to share a diverse set of ~10,000 real-time variables, including arrays, across more than 100 computers, with information being published every 10 ms in some cases. SMA-X is open-source, and is freely available to everyone through a set of public GitHub repositories, including C/C++ and Python3 libraries to allow integration with observatory applications. A set of command-line tools provide access to the database from the POSIX shell and/or from any scripting language, and we also provide a configurable system daemon for archiving the observatory state at regular intervals into a time-

This paper introduces a microservices architecture for the purpose of enhancing the flexibility and performance of an airline reservation system. The architectural design incorporates Redis cache technologies, two different messaging systems (Kafka and RabbitMQ), two types of storages (MongoDB, and PostgreSQL). It also introduces authorization techniques, including secure communication through OAuth2 and JWT which is essential with the management of high-demand travel services. According to selected indicators, the architecture provides an impressive level of data consistency at 99.5% and a latency of data propagation of less than 75 ms allowing rapid and reliable intercommunication between microservices. A system throughput of 1050 events per second was achieved so that the acceptability level was maintained even during peak time. Redis caching reduced a 92% cache hit ratio on the database thereby lowering the burden on the database and increasing the speed of response. Further improvement of the systems scalability was done through the use of Docker and Kubernetes which enabled services to be expanded horizontally to cope with the changes in demand. The error rates were very low,