This guide demonstrates multiple ways to gather Nginx access and error logs—directly with Filebeat to Elasticsearch, via Filebeat to Logstash then Elasticsearch, and using rsyslog to forward logs to Logstash—providing step‑by‑step configurations, code snippets, and visual illustrations for each method.
The article presents a comprehensive performance benchmark of RedisJSON (with RediSearch) against MongoDB and ElasticSearch, showing RedisJSON’s superior write, read, and mixed‑workload throughput and latency across isolated and mixed scenarios, supported by detailed YCSB test results and analysis.
This article provides a step‑by‑step tutorial on integrating SpringBoot with Spring Data Elasticsearch, covering required dependencies, configuration classes, entity mapping, repository interfaces, service implementations, controller endpoints, and testing instructions, complete with runnable code examples.
This article explains the principles behind Elasticsearch searchable snapshots, details the DataTier model and node role optimizations, and provides a full practical walkthrough—including cluster setup, COS repository creation, ILM policy configuration, index templates, mounting strategies, and performance considerations—using ES 7.14.2.
Elasticsearch is a distributed, schemaless search engine built on Lucene that stores JSON documents in sharded indexes, uses immutable segments and merges, provides a flexible Query DSL with aggregations and relevance scoring, and executes distributed query‑then‑fetch searches with features like scrolling, optimistic locking, and zero‑downtime reindexing.
The article explains Elasticsearch 7.14’s production‑ready searchable snapshot feature, detailing tiered node roles, allocation preferences, full and partial mount types, and provides a step‑by‑step Tencent Cloud COS setup—including license activation, repository creation, ILM policy, index template and automatic mounting—to achieve cost‑effective, large‑scale data storage.
This article explains how Elasticsearch uses inverted indexes, term dictionaries, and compression techniques like FOR and Roaring Bitmaps to enable rapid full‑text search, contrasting its approach with traditional relational databases and offering practical indexing tips for large‑scale applications.
The report presents extensive YCSB‑based performance benchmarks showing that RedisJSON (with RediSearch) outperforms MongoDB and ElasticSearch by large margins in isolated writes, isolated reads, mixed workloads, and latency across a variety of operation mixes on identical AWS hardware.
This article details JD Retail's strategic "Nirvana" product‑selection platform, describing the technical challenges of handling billions of items and hundreds of tags, and presenting a dual‑engine solution using ClickHouse and Elasticsearch with Spark‑driven data pipelines to achieve fast filtering, multidimensional analytics, and efficient storage.
The Smart Matching Engine for Haolo’s ride‑sharing service ingests driver and passenger orders via Kafka‑Flink pipelines into Elasticsearch, then applies multi‑stage matching—nearby search, itinerary‑based filtering using ETA, angle, distance, route‑similarity and shared‑mileage calculations—and finally ranks results with evolving pre‑sorting and algorithmic models, including PMML and deep‑learning, to optimize driver‑passenger pairing.
This article explains Elasticsearch’s cluster and data layer architecture, covering nodes, indices, shards, replicas, mixed and tiered deployment models, storage options, and the trade‑offs of different distributed system designs for scalable systems.
This article explains how Elasticsearch leverages Lucene's inverted index, term dictionaries, and posting list compression methods such as FST and Frame‑of‑Reference to achieve fast, accurate, and tolerant search performance, while also discussing practical implementation details and examples.
This article explains how Elasticsearch leverages Lucene’s inverted index, detailing the structure of term dictionaries, postings lists, compression techniques like Frame‑of‑Reference and Roaring Bitmaps, and query optimizations such as filter caches and skip‑list intersections to achieve fast, memory‑efficient search.
This article explains why centralized log collection is essential for multi‑environment services, introduces Graylog’s architecture, details Filebeat’s role and workflow, provides configuration examples, and walks through Docker‑based deployment of both Filebeat and Graylog for robust log management.
This article explains why MySQL queries become slow, how index design and lock contention affect performance, and when to complement MySQL with Elasticsearch for full‑text search or HBase for write‑intensive workloads, providing practical optimization strategies and architectural guidance.
This guide walks through creating a file upload service that indexes Word, PDF, and TXT files in Elasticsearch, uses an ingest‑attachment pipeline to extract text, configures Chinese IK analyzers for precise keyword search, and demonstrates Java code for indexing, querying, and highlighting results.
This article explains how to integrate Elasticsearch into a Spring Boot project, covering version compatibility, supported access methods, dependency setup, configuration, entity mapping with annotations, repository creation, and unit testing to verify successful indexing and retrieval.
This guide explains how to use Filebeat to ship logs to Graylog, covering Filebeat's architecture, configuration files, deployment options with Docker or native packages, Graylog's components and pipelines, and step‑by‑step Docker‑compose setup for a scalable centralized logging solution.
This tutorial shows how to quickly construct a streaming ETL pipeline that captures changes from MySQL and PostgreSQL using Flink CDC, enriches order data with product and shipment information, and writes the results into Elasticsearch for real‑time visualization in Kibana.
This article explains why ID mapping is essential for breaking data silos, describes traditional address cleaning and DSP scenarios, and details a graph‑based, six‑step process that builds a unified One‑ID dictionary to enable comprehensive user profiling and analytics in big‑data environments.
This article introduces three widely used system architecture designs—single‑database single‑application, content distribution with CDN/OSS, and read‑write separation with master‑slave databases and Elasticsearch—detailing their structures, advantages, disadvantages, and typical usage scenarios.
This article provides a detailed guide on Elasticsearch index design, covering index settings, shard and replica planning, mapping strategies, complex types, lifecycle management, template usage, and practical best‑practice recommendations for large‑scale log data clusters.
This article introduces Elasticsearch as a distributed, high‑performance search engine, explains its cluster architecture, node roles, shard and replica mechanisms, write and refresh processes, and outlines search handling techniques such as pagination, scroll and search‑after with practical curl examples.
This article explains why MySQL queries become slow, how index design and common pitfalls affect performance, introduces MDL locks and large‑table strategies, and then compares Elasticsearch and HBase as complementary solutions for fast search and massive data storage.
This article explains the core concepts of Elasticsearch’s distributed architecture—including nodes, indices, shards, replicas—and compares mixed and tiered deployment models, while also discussing data storage strategies, replica benefits, and the trade‑offs of local‑file versus shared‑storage distributed systems.
This article explains the core concepts of Elasticsearch—including nodes, indices, shards, and replicas—covers its cluster and data‑layer architectures, compares mixed and tiered deployment models, and discusses the advantages and drawbacks of replica‑based distributed storage systems.
This article explains how Elasticsearch uses inverted indexes, term dictionaries, and advanced compression techniques like Frame of Reference and Roaring Bitmaps to enable rapid, scalable search over massive datasets, contrasting its approach with traditional relational database queries and detailing practical optimization tips.
This article explains how Elasticsearch implements near real-time search by using immutable inverted indexes, segment merging, shard distribution, and a translog for durability, detailing the challenges of persistence, disk I/O, and data recovery in a distributed environment.
This article details how a rapidly growing e‑commerce platform migrated from a single MySQL instance to a 16‑shard architecture using Sharding‑Jdbc, introduced consistent‑hashing to mitigate data skew, leveraged ES+HBase for multi‑dimensional queries, and implemented zero‑downtime migration strategies such as dual‑write and Canal replication.
This tutorial explains the principles of distributed tracing, why it is needed for microservice architectures, and provides step‑by‑step instructions for adding Spring Cloud Sleuth and Zipkin—including Maven dependencies, configuration, Docker deployment, and Elasticsearch persistence—to a Spring Cloud project.
This guide walks through setting up a complete log‑collection and visualization pipeline—including SpringBoot log4j2 configuration, Kafka broker creation, Filebeat forwarding, Logstash processing, and Kibana dashboard setup—so you can capture, ship, and analyze application logs in real time.
This article explains how Elasticsearch implements near real-time search by using immutable inverted indexes, segment merging, sharding, and a translog for durability, detailing the challenges and solutions behind its distributed full‑text search architecture.
This guide walks you through setting up a complete log‑collection and visualization pipeline—preparing servers, configuring a SpringBoot application with Log4j2, routing logs to Kafka, harvesting them with Filebeat, processing with Logstash, and finally visualizing in Elasticsearch and Kibana.
This guide walks through three practical methods for harvesting Nginx access and error logs—directly with Filebeat to Elasticsearch, via Filebeat‑Logstash‑Elasticsearch pipeline, and using Rsyslog to forward logs to Logstash—complete with configuration snippets and visual illustrations.
This article explains the architecture of Elasticsearch clusters, covering node roles, index, shard and replica concepts, deployment models, data storage mechanisms, and compares two distributed system designs—local‑file‑system and shared‑file‑system—highlighting their advantages and trade‑offs.
This article explains how Elasticsearch uses inverted indexes, term dictionaries, and postings lists—along with compression techniques like Frame of Reference and Roaring Bitmaps—to achieve fast, memory‑efficient search queries, and provides practical tips for optimizing indexing and query performance.
This article details the design and implementation of a high‑performance, highly available membership system for the merged Tongcheng‑eLong platform, covering Elasticsearch dual‑center clusters, traffic‑isolated three‑cluster architecture, deep ES optimizations, Redis caching and dual‑center clusters, MySQL dual‑center partitioning, migration strategies, and future fine‑grained flow‑control and degradation measures.
This article explains how Elasticsearch uses inverted index structures—including term dictionaries, term indexes, and postings lists—combined with compression methods like Frame‑of‑Reference and Roaring Bitmaps to achieve fast search, efficient storage, and effective union queries compared to traditional relational databases.
This article explains Elasticsearch's core mechanisms for indexing, querying, and searching data, covering the roles of coordinating nodes, primary and replica shards, refresh cycles, translog, commit/flush processes, and the underlying Lucene inverted index.
This guide walks through setting up a complete log‑collection and visualization pipeline—preparing servers, configuring a SpringBoot project with Log4j2, deploying Kafka, installing Filebeat, creating Logstash pipelines, and visualizing logs in Elasticsearch and Kibana—so you can monitor application logs in real time.
This article extends previous Elasticsearch cheat sheets with practical commands and step‑by‑step instructions for shard allocation, replica adjustment, cluster settings, slow‑log configuration, mapping routing, force merge, bulk writes, refresh intervals, translog durability, heap sizing, disk‑space monitoring, and troubleshooting strategies.
This article explains Elasticsearch's core concepts—nodes, indices, shards, and replicas—describes the indexing flow, compares mixed and layered deployment models, and discusses data‑layer architectures, helping readers choose the appropriate cluster design for reliability and scalability.
This article analyzes Elasticsearch’s distributed architecture, explaining core concepts such as nodes, indices, shards and replicas, comparing mixed and layered deployment models, and discussing data‑layer design choices and their trade‑offs in modern distributed storage and search systems.
This guide shows how to configure a Spring Boot 2.3.12 application with Elasticsearch 7.8, create index templates, implement a reusable bulk operator service, and test bulk insertion of task data into dynamically named indices, ensuring automatic index creation and proper mapping.
This tutorial walks through the end‑to‑end setup of a logging pipeline that starts with a SpringBoot application using Log4j2, forwards logs to Kafka, collects them with Filebeat, processes them via Logstash, and finally visualizes them in Elasticsearch and Kibana, covering server preparation, configuration files, and essential code snippets.
This article compares Elasticsearch and ClickHouse in terms of architecture, query capabilities, and performance for log analytics, presenting test setups, Docker‑compose configurations, query examples, and benchmark results that show ClickHouse generally outperforms Elasticsearch in most basic query scenarios.
This guide demonstrates how to set up Spring Boot 2.3.12 with Elasticsearch 7.8, configure separate JPA and Elasticsearch repositories, and implement two batch insertion methods—using ElasticsearchRestTemplate and repository saveAll—while measuring performance, showing comparable execution times for inserting 10,000 records.
This article details Qunar's design and implementation of a data synchronization platform that aggregates MySQL data into Elasticsearch, supporting parallel ES5.x/7.x clusters, hot‑swap upgrades, flexible scaling, and end‑to‑end consistency with high availability mechanisms.
This article demonstrates how to construct a full‑text site search solution by selecting Elasticsearch as the search engine, Spring Boot for the backend service, and the IK analyzer for Chinese tokenization, covering environment setup, project architecture, key code implementations, UI pages, and a concise conclusion on the effectiveness of the approach.
This step‑by‑step guide explains how to set up the open‑source ELK stack (Elasticsearch, Logstash, Kibana) on an Ubuntu machine, configure Logstash as shipper and indexer, integrate Spring Boot and Nginx logs, and run the whole platform as a background service using Supervisor.
This article explains why MySQL queries become slow, how proper indexing, MDL locks, sharding, read‑write separation, and large‑table strategies can improve performance, and also introduces Elasticsearch and HBase as complementary technologies for search and high‑write workloads.
This article introduces how to build a global code search system called 'Qianxun' from scratch, covering its background, architecture, core technologies, and future prospects.
This article provides a comprehensive overview of Elasticsearch, covering its fundamental architecture, key concepts such as indices, shards and replicas, the complete write and search workflows, consistency mechanisms, master node election, and practical performance‑tuning recommendations for large‑scale deployments.
This tutorial explains how to set up an open‑source ELK (Elasticsearch, Logstash, Kibana) stack for real‑time log collection, parsing, and visualization, covering component installation, Shipper/Indexer configuration, Grok pattern creation, Nginx integration, and background service management with Supervisor.
This article explains how Elasticsearch achieves rapid search by using inverted indexes, detailing the structure of posting lists, term dictionaries, compression methods like Frame‑of‑Reference and Roaring Bitmaps, and how these techniques enable efficient union queries and filter caching.
This tutorial walks through preparing servers, configuring a SpringBoot project with Log4j2, setting up Kafka, Filebeat, and Logstash to collect and parse logs, and finally visualizing them in Elasticsearch and Kibana, providing a full end‑to‑end logging solution.
This article explains how Elasticsearch leverages Lucene's inverted index, term dictionary, term index, and compression methods such as Frame‑of‑Reference and Roaring Bitmaps to enable rapid search, efficient storage, and fast set operations for large‑scale data retrieval.
This article explains how Elasticsearch achieves fast data retrieval by comparing it with traditional relational databases, detailing search engine fundamentals, the structure of Lucene's inverted index—including term dictionaries, postings lists, and term indexes—and the compression techniques such as Frame of Reference and Roaring Bitmaps that optimize storage and query performance.
This article explains three practical methods for assigning default values in Elasticsearch—using ingest pipelines, update‑by‑query scripts, and pipeline scripts—while also addressing how to maintain create_time and update_time fields in a way similar to relational databases.
The first Shenzhen Elasticsearch meetup on August 21, 2021, jointly hosted by the ES Chinese community and Tencent Cloud, gathered experts from Tencent, Tapdata, ByteDance and Vivo to showcase rapid community growth, compression‑encoding optimizations, real‑time ES‑MongoDB data fusion, custom kernel extensions, large‑scale cluster practices, and concluded with extensive Q&A and networking.
This article shares practical lessons from three iterations of a data platform, focusing on Elasticsearch and Lucene optimizations that enable cross‑month queries, year‑long data export, and sub‑second query responses for tables handling billions of rows per day.
This guide presents practical best‑practice configurations for Elasticsearch clusters in production, covering JVM heap sizing, CPU thread‑pool tuning, optimal shard counts, replica strategies, hot‑warm node architecture, node role settings, common troubleshooting tips, cache handling, refresh intervals, and essential monitoring APIs.
In July, security researcher Bob Diachenko uncovered an exposed Elasticsearch cluster leaking nearly two million personal records—including passport details and no‑fly indicators—highlighting the severe impact of unsecured Elasticsearch deployments and offering recommendations to prevent future breaches.
This guide walks you through creating, viewing, updating, and deleting Elasticsearch indices, documents, and mappings using RESTful HTTP methods, complete with example requests, JSON bodies, and visual illustrations for each operation.
This article explores comprehensive domain interface design in Java backend development, showing how to model domain objects as interfaces, implement repositories for JPA, MyBatis, and Elasticsearch, handle associations with JPA annotations, and apply these patterns in the open‑source Mallfoundry e‑commerce platform.
This article explains why MySQL queries become slow, how proper indexing and index‑pushdown can improve performance, discusses common index‑failure causes, and then introduces Elasticsearch and HBase as complementary search and storage solutions for large‑scale data, including practical usage tips and architectural considerations.
This article describes how a top architect optimized Elasticsearch for handling billions of records, covering Lucene fundamentals, index and shard design, DocValues, query performance tuning, bulk indexing strategies, hardware considerations, and testing methods to achieve sub‑second query responses across multi‑year data ranges.
The Shenzhen Elasticsearch technical event, co‑hosted by the Elastic Chinese community and Tencent Cloud, presented practical sessions on optimizing the Elastic Stack for search, real‑time analytics, logging, security and APM, featuring compression encoding, MongoDB fusion, ByteDance extensions, cost‑effective log storage, Lucene indexing, cross‑cluster replication, vector engine integration, and large‑scale case studies from Tencent, Tiptop Data and vivo.
The article details Qunar's data synchronization platform that aggregates MySQL data into Elasticsearch, covering its architecture, component choices, ES5‑to‑ES7 migration, hot‑plugging, reindexing, high‑availability design, consistency guarantees, operational optimizations, and future roadmap.
This step‑by‑step guide shows how to configure a Spring Boot application with Log4j2, ship logs via Filebeat to Kafka, process them with Logstash, and visualize the data in Elasticsearch and Kibana, covering server setup, Maven dependencies, configuration files, and deployment commands.
This article explains why MySQL queries become slow—covering index misuse, MDL locks, flush waits, row locks, and large‑table bottlenecks—then introduces ElasticSearch’s inverted‑index architecture and HBase’s column‑family storage, offering practical tips for performance tuning and hybrid solutions.
This article provides a comprehensive introduction to Elasticsearch, covering its underlying Lucene-based inverted index, data types, shard routing, cluster roles, discovery mechanisms, refresh and translog handling, segment merging, and practical performance and JVM tuning tips for building scalable, near‑real‑time search systems.
This article examines JD Daojia’s after‑sale platform, detailing its three‑tier architecture, distributed‑lock mechanisms for multi‑endpoint requests, complex promotion‑aware split‑data handling, Elasticsearch synchronization, combined‑order logistics, and rigorous refund validation to ensure accurate and performant service delivery.
This article explains why MySQL queries become slow, how proper indexing—including B+‑tree, left‑most prefix, index push‑down, and covering indexes—can improve performance, outlines common causes of index failure, and then introduces ElasticSearch search capabilities and HBase column‑family storage as complementary solutions for large‑scale data handling.
This tutorial explains how to set up a real‑time ELK log analysis platform, covering the architecture of Elasticsearch, Logstash and Kibana, detailed installation commands, configuration for Spring Boot and Nginx logs, and how to run the stack continuously with Supervisor.
Elastic's recent client‑library changes restrict connections to AWS‑hosted OpenSearch and older Elasticsearch versions, prompting AWS to fork multiple official clients and sparking a heated debate over open‑source licensing and developer freedom.
This article presents a comprehensive guide on optimizing Elasticsearch for large‑scale data platforms, covering Lucene fundamentals, index and shard architecture, doc‑values usage, routing strategies, practical performance‑tuning techniques, and real‑world testing results to achieve sub‑second query responses on billions of records.
This article outlines a backend messaging center architecture that targets 10,000 messages per second inbound via RocketMQ, 1,000 messages per second outbound to third‑party platforms, and ensures 100% high availability using Spring Cloud Gateway, Kubernetes, Elasticsearch, and related technologies.
This article introduces the most important and still mainstream components of the big data ecosystem—including Hadoop’s storage and compute framework, Hive data warehouse, HBase NoSQL database, Spark unified engine, Kafka messaging platform, and Elasticsearch search engine—explaining their core concepts, architectures, and typical use cases.
This article compares Elasticsearch and ClickHouse for log analytics, detailing their architectures, presenting a Docker‑Compose test setup that streams synthetic syslog data to both stacks, executing equivalent queries, measuring response times, and concluding that ClickHouse generally delivers faster performance, especially for aggregations.
This article explains the complete technical design of an e‑commerce coupon system, covering data models, product‑page exposure, coupon claiming rules, high‑concurrency caching, MQ‑driven expiration handling, order‑time coupon calculation, redemption flow, and search‑based coupon indexing.
This article explains the design and evolution of Baidu's word‑list service for content moderation, covering its background, multi‑layer architecture, management platform, strategy loading, matching workflow, performance optimizations for large texts, and future enhancements such as special‑character support and per‑business‑line deployment.
This article presents a practical, out‑of‑the‑box ELK‑based solution for private deployments, detailing design principles, rapid one‑click setup with Jenkins, component choices, log and metric collection using Beats, alerting with ElastAlert, and automated Kibana dashboard configuration.
This guide walks you through installing and configuring the ELK stack—Elasticsearch, Logstash, and Kibana—on Ubuntu, setting up Logstash shipper and indexer pipelines, integrating Spring Boot and Nginx logs, and managing the services with Supervisor for reliable, real‑time log analysis.
This article compares Elasticsearch and ClickHouse for log analytics, detailing their architectures, node roles, query capabilities, and performance through Docker‑Compose stacks and synthetic syslog data, concluding that ClickHouse generally outperforms Elasticsearch in speed and aggregation efficiency.
This article compares Elasticsearch and ClickHouse as log analytics solutions, detailing their architectures, deployment configurations, query capabilities, and performance benchmarks across various query types, and demonstrates that ClickHouse generally outperforms Elasticsearch in speed and aggregation efficiency.
This article compares ClickHouse and Elasticsearch for API log storage, detailing development activity, schema handling, query performance, statistical functions, MySQL integration, new features, and practical drawbacks, while providing concrete SQL examples and migration tips.
This article explores common causes of MySQL slow queries such as index misuse and lock contention, explains indexing strategies like index pushdown and covering indexes, and then compares Elasticsearch and HBase as complementary solutions for large‑scale search and write‑intensive workloads, offering practical tips for performance optimization.
This article outlines the technical, business, and product goals for a messaging center, presents a prototype and functional requirements, evaluates RocketMQ and Elasticsearch as core technologies, and details the architectural design, underlying frameworks, and DevOps strategies—including Spring Cloud Gateway, Kubernetes, and Docker—to achieve 10,000 msg/s upstream throughput, 1,000 msg/s downstream delivery, and 100 % high availability.
This article explains how Elasticsearch leverages inverted indexes, term dictionaries, and advanced compression techniques like Frame of Reference and Roaring Bitmaps to enable rapid full‑text search, covering the underlying concepts, data structures, and query optimizations essential for high‑performance search applications.
This article explains how Elasticsearch achieves fast search by using inverted indexes, detailing the structure of posting lists, term dictionaries, term indexes, and compression methods such as Frame of Reference and Roaring Bitmaps, as well as techniques for efficient union and intersection queries.
This article explains Loki's advantages over Elasticsearch, including low indexing overhead, concurrent query processing with caching, seamless integration with Prometheus and Grafana, detailed architecture components, installation steps, label handling, high‑cardinality challenges, and best practices for efficient log management.
This article examines common causes of slow queries in MySQL, presents practical indexing techniques, and extends the discussion to performance considerations in ElasticSearch and HBase, offering combined solutions such as sharding, read/write splitting, and using ES as a search layer for large‑scale systems.
Zhang Liang shares Didi's four-stage journey of constructing and governing large‑scale open‑source big‑data engine services—including engine selection, hardware sizing, PaaS platform building, proxy architecture, and governance—highlighting practical challenges, solutions, and ROI‑driven best practices for Kafka, Elasticsearch, Flink, and related technologies.
This article explains why large‑scale Elasticsearch clusters may miss log entries during keyword searches, dives into the fundamentals of inverted indexes and tokenization, and demonstrates practical index‑time and query‑time tokenizer optimizations—including custom analyzers for English and Chinese—to dramatically improve search recall and precision.
This tutorial walks through creating a Flask‑based search engine powered by Elasticsearch, covering configuration files, logging setup, blueprint routing, pagination, detail views, and deployment options using Flask‑Script and Gunicorn, with complete code examples and a GitHub reference.
This guide walks through installing the Python Elasticsearch client, building a reusable class with CRUD methods, importing data from MongoDB, writing a simple Baidu Baike crawler, and scaling the workflow with Celery and Flask for a complete search‑engine pipeline.
This tutorial walks you through why ELK is essential for large‑scale log collection, shows a simple architecture diagram, and provides detailed installation and configuration steps for Elasticsearch, Logstash, and Kibana on both macOS and Linux, including common pitfalls and Java integration.
Learn what Elasticsearch is, its core concepts, and how to integrate it into a Django project, including installation, configuration, defining indices and documents, populating data, and performing searches, all illustrated with code examples for seamless full‑text search capabilities.
This article explains Elasticsearch's write, read, and search mechanisms, the role of coordinating nodes, primary and replica shards, refresh and commit cycles, Lucene's inverted index, and how data becomes searchable in near‑real‑time.
