{"id":32939,"date":"2026-04-09T11:52:49","date_gmt":"2026-04-09T18:52:49","guid":{"rendered":"https:\/\/www.pingcap.com\/?p=32939"},"modified":"2026-04-10T11:55:01","modified_gmt":"2026-04-10T18:55:01","slug":"s3-new-network-cloud-object-storage-database-architecture","status":"publish","type":"post","link":"https:\/\/www.pingcap.com\/ko\/blog\/s3-new-network-cloud-object-storage-database-architecture\/","title":{"rendered":"S3 is the New Network: Rethinking Data Architecture for AI Agents"},"content":{"rendered":"<p><em>Editor\u2019s note: This post originally appeared on The New Stack and is republished with permission. The original version is available\u00a0<a href=\"https:\/\/thenewstack.io\/tidb-x-open-source-database\/\" target=\"_blank\" rel=\"noreferrer noopener\">\uc5ec\uae30<\/a>.<\/em><\/p>\n\n\n\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"Key_Takeaways\"><\/span><strong>Key Takeaways<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<ul class=\"wp-block-list\">\n<li>S3&#8217;s durability and global availability remove the need to co-locate data with compute.<\/li>\n\n\n\n<li>Decoupled storage enables ephemeral clusters, event-driven workflows, and automatic tiering.<\/li>\n\n\n\n<li>TiDB X uses S3 as its shared backend for independent scaling and faster recovery.<\/li>\n\n\n\n<li>Object storage-first architecture matches the elastic, on-demand needs of AI agents.<\/li>\n<\/ul>\n<\/blockquote>\n\n\n\n<p>For decades, database designers have built <a href=\"https:\/\/www.pingcap.com\/ko\/blog\/why-distributed-sql-databases-elevate-modern-app-dev\/\">distributed databases<\/a> around the assumption that storage must live close to compute.<\/p>\n\n\n\n<p>The farther data travels over the network, the reasoning goes, the greater the potential for delay. Local RAID (redundant array of independent disks) arrays, network-attached storage (NAS), and cluster file systems keep data close, making it quick and easy to access.&nbsp;<\/p>\n\n\n\n<p>But in a distributed system, keeping the entire data store close to compute makes scaling slow, cumbersome, and expensive. Each time you replicate a node or cluster, you must replicate its associated data as well.<\/p>\n\n\n\n<p>It isn\u2019t ideal, but until recently, there wasn\u2019t any reasonable alternative. Databases had to scale. Teams had to meet service-level agreements (SLAs). Wide-area networks weren\u2019t reliable enough to support high-performance databases at scale. Database designers accordingly spent a great deal of energy solving problems related to coordination, consistency, and replication logic.<\/p>\n\n\n\n<p>But imagine things were different. What if they didn\u2019t have to worry about the network, where their data lived, or how to get it from Point A to Point B? How would they design a database then?<\/p>\n\n\n\n<p>That\u2019s the intriguing question raised by the advent of cloud object storage services like AWS S3, Google Cloud Storage, and Microsoft Blob Storage.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"What_is_Cloud_Object_Storage\"><\/span>What is Cloud Object Storage?<span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p>The structure of cloud object storage services couldn\u2019t be simpler. They\u2019re essentially giant heaps of data, accessed via an API, through key\/value pairings.<\/p>\n\n\n\n<p>Their unlimited storage capacity and their \u201ceverywhere\u201d availability make them revolutionary. They can hold billions of records \u2014 images, logs, training data, whatever you need \u2014 and crucially, they can make every one of those records available to compute anywhere in the world, at any level of workload.<\/p>\n\n\n\n<p>S3 is extremely reliable. AWS designed S3 for 11 nines of durability (that&#8217;s 99.999999999%) and 99.99% availability, and it replicates data automatically across Amazon&#8217;s regional facilities. This means data on S3 is extremely safe and highly available without the need to manage physical disks or replication.<\/p>\n\n\n\n<p>In addition, S3 scales seamlessly. There are no fixed volumes. No need for capacity planning. You can store practically unlimited data, and performance scales with parallel access rather than a single-server bottleneck. These guarantees free architects from worrying about low-level storage failures, capacity, and edge cases involving consistency.\u00a0<\/p>\n\n\n\n<p>In short, cloud object storage provides a highly durable, always-on, strongly-consistent single source of truth. It\u2019s not as fast as local storage, but it doesn\u2019t have to be. What services like S3 lack in sheer speed, they more than make up for in reliability and ease of maintenance. Instead of worrying about shards, segmentation, and software-defined networks, a database can simply retrieve data with confidence that it will be delivered in a reasonable amount of time.&nbsp;<\/p>\n\n\n\n<p>What this means is that for the next generation of distributed databases, cloud object storage will, for all intents and purposes, <em>be <\/em>the network.&nbsp;<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"Architectural_Patterns_Emerging_Around_Object_Storage\"><\/span>Architectural Patterns Emerging Around Object Storage<span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p>Building on cloud object storage enables several architectural patterns that were previously impractical.<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>Ephemeral compute clusters:<\/strong> Keeping object storage separate from compute makes it easier to spin up clusters temporarily for a specific job and tear them down afterward. This is especially useful for <a href=\"https:\/\/www.pingcap.com\/ko\/blog\/how-manus-1-5-uses-tidb-x-to-let-agents-ship-full-stack-apps-at-scale\/\">AI agents, which often construct temporary databases<\/a> to accomplish tasks. Compute can be spun up at will without the overhead of data replication.<br><\/li>\n\n\n\n<li><strong>Event-driven workflows:<\/strong> The arrival of a new object in S3 can trigger a Lambda function, start a training job, or notify downstream consumers. This sort of workflow would be impractical in a system with highly replicated data, but it\u2019s trivial when data is centralized in a single store.<br><\/li>\n\n\n\n<li><strong>AI and ML pipelines:<\/strong> Many distributed machine learning workflows benefit from a centralized object storage data store. Training datasets, feature stores, model checkpoints, and experiment logs all commonly live in object stores. Frameworks like TensorFlow, PyTorch, and SageMaker are designed to stream data directly from object storage.<br><\/li>\n\n\n\n<li><strong>Tiering storage at large scale<\/strong>: Databases often classify data as either in-demand (\u201chot\u201d) or rarely accessed (\u201ccold\u201d). Hot data is stored on high-speed flash storage, while cold data is stored on a more cost-efficient spinning disk. Provisioning hot and cold storage normally requires manual intervention and careful capacity planning. But with cloud object storage, the database can automatically handle tiering, shuffling data between the object store and the high-speed cache based on demand. The availability and infinite capacity of the object store make planning unnecessary.<\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"Example_TiDB_X\"><\/span>Example: TiDB X<span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p>Now let\u2019s see how these capabilities translate into a real-world design. PingCAP uses cloud object storage as the foundation for <a href=\"https:\/\/www.pingcap.com\/ko\/blog\/introducing-tidb-x-a-new-foundation-distributed-sql-ai-era\/\">TiDB X<\/a>, the latest version of our popular open source distributed SQL database, TiDB.<\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"565\" src=\"https:\/\/static.pingcap.com\/files\/2026\/04\/10111323\/image-1024x565.png\" alt=\"TiDB X's built-in cloud object storage architecture.\" class=\"wp-image-32940\" srcset=\"https:\/\/static.pingcap.com\/files\/2026\/04\/10111323\/image-1024x565.png 1024w, https:\/\/static.pingcap.com\/files\/2026\/04\/10111323\/image-300x166.png 300w, https:\/\/static.pingcap.com\/files\/2026\/04\/10111323\/image-768x424.png 768w, https:\/\/static.pingcap.com\/files\/2026\/04\/10111323\/image-1536x848.png 1536w, https:\/\/static.pingcap.com\/files\/2026\/04\/10111323\/image.png 1600w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<p class=\"has-text-align-center\"><em>Figure 1. TiDB X\u2019s architecture with built-in object storage.<\/em><\/p>\n\n\n\n<p>As shown in the diagram above, TiDB X <a href=\"https:\/\/www.pingcap.com\/ko\/tidb\/\">fully separates compute and storage<\/a>, using S3 for the shared backend. Compute nodes scale independently up and down. Fast local caches and Raft ensure consistency and low-latency access for hot data. Instead of keeping the entire data store close by, TiDB X keeps only the most active data near compute. TiDB X monitors query patterns, latency targets, and data characteristics, then reshapes itself in response to demand.\u00a0<\/p>\n\n\n\n<p>Its object storage-based architecture streamlines recovery and backup processes. By using S3 for primary data persistence, TiDB X reduces the overhead of traditional backup maintenance, enabling significantly faster completion times. This design also mitigates the impact of node failures: since local state functions primarily serve as a cache for durable, replicated storage, a failed instance can be replaced by retrieving its required state directly from object storage to resume operations.<\/p>\n\n\n\n<p>From an operational perspective, cloud object storage makes TiDB X both highly adaptable and extremely cost-efficient. Its autoscaler responds not just to preset infrastructure thresholds, but to contextual signals like query patterns, latency targets, and data types. This enables it to reshape its resources in real time to address different tasks.<\/p>\n\n\n\n<p>In sum, by building atop AWS\u2019s high-performance object data store, TiDB X demonstrates how a <a href=\"https:\/\/www.pingcap.com\/ko\/tidb\/cloud\/\">cloud database can achieve elasticity, performance, and simplicity<\/a> without sacrificing consistency or scale.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\"><span class=\"ez-toc-section\" id=\"S3_as_the_Communication_Fabric\"><\/span>S3 as the Communication Fabric<span class=\"ez-toc-section-end\"><\/span><\/h2>\n\n\n\n<p>Keeping large relational data stores close to compute resources has always been a compromise. It was an expensive solution to a problem created by the limitations of traditional networking.&nbsp;<\/p>\n\n\n\n<p>With architectures like TiDB, we see that the sheer power and scale of services like S3 have made the old workarounds unnecessary. They\u2019ve rendered traditional architectures increasingly obsolete. More than that, they\u2019ve enabled practices, such as ephemeral compute, suited to a world where users are more likely to be AI agents than humans.<\/p>\n\n\n\n<p>As AI reshapes business organizations and best practices, the database itself is changing form. In large part, it\u2019s services like S3 that are making that shift possible. By making data placeless, ubiquitous, and effortlessly accessible<em>, <\/em>cloud object storage is overturning the assumptions that once guided database design. The result will be databases that are more flexible and resilient&nbsp; \u2014 ones that are simpler to manage and scale almost effortlessly.<\/p>\n\n\n\n<p><em>TiDB X is built from the ground up on cloud object storage. <a href=\"https:\/\/www.pingcap.com\/ko\/blog\/tidbx-origins-architecture\/\">Explore the architecture behind it<\/a> \ub610\ub294 <a href=\"https:\/\/tidbcloud.com\/free-trial\/?__hstc=86493575.783064bfcc857ae1a573df16c96a21a4.1767977986672.1775842627511.1775846751398.357&amp;__hssc=86493575.2.1775846751398&amp;__hsfp=c9b8f6b9119f97b0d0459ed29a93bd43&amp;_gl=1*1601w2d*_gcl_au*MTU1MzYwNTQ4My4xNzc1NzYzNzQ1*_ga*MjUyOTQyMTU0LjE3Njc5Nzc5ODQ.*_ga_9FRXHHPYVY*czE3NzU4NDY3NTEkbzM4OCRnMSR0MTc3NTg0NjgyMyRqNjAkbDAkaDA.*_ga_3JVXJ41175*czE3NzU4NDI2MjYkbzM5MCRnMSR0MTc3NTg0NjgyMyRqNjAkbDAkaDIxMjU2NjA5MTA.*_ga_ZEL0RNV6R2*czE3NzU4NDI2MjYkbzM0NiRnMSR0MTc3NTg0NDg0MCRqNTYkbDAkaDA.&amp;website_referrer_url=https:\/\/pingcap.zoom.us\/\">try TiDB Cloud for free<\/a> to see it in action.<\/em><\/p>","protected":false},"excerpt":{"rendered":"<p>Editor\u2019s note: This post originally appeared on The New Stack and is republished with permission. The original version is available\u00a0here. For decades, database designers have built distributed databases around the assumption that storage must live close to compute. The farther data travels over the network, the reasoning goes, the greater the potential for delay. Local [&hellip;]<\/p>\n","protected":false},"author":50,"featured_media":32948,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"ub_ctt_via":"","footnotes":""},"categories":[145],"tags":[483,147,482,9,440],"class_list":["post-32939","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-thought-leadership","tag-ai-agents","tag-distributed-sql","tag-object-storage","tag-scalability","tag-tidb-x"],"acf":[],"featured_image_src":"https:\/\/static.pingcap.com\/files\/2026\/04\/10114857\/Blog-Feature.png","author_info":{"display_name":"Max Liu","author_link":"https:\/\/www.pingcap.com\/ko\/blog\/author\/max-liu\/"},"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v26.9 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>Cloud Object Storage Architecture: S3 Is the New Network<\/title>\n<meta name=\"description\" content=\"Explore the cloud object storage architecture reshaping databases and how TiDB X uses S3 for elastic scale, fast recovery, and AI workloads.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/thenewstack.io\/tidb-x-open-source-database\/\" \/>\n<meta property=\"og:locale\" content=\"ko_KR\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Cloud Object Storage Architecture: S3 Is the New Network\" \/>\n<meta property=\"og:description\" content=\"Explore the cloud object storage architecture reshaping databases and how TiDB X uses S3 for elastic scale, fast recovery, and AI workloads.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/thenewstack.io\/tidb-x-open-source-database\/\" \/>\n<meta property=\"og:site_name\" content=\"TiDB\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/facebook.com\/pingcap2015\" \/>\n<meta property=\"article:published_time\" content=\"2026-04-09T18:52:49+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2026-04-10T18:55:01+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/static.pingcap.com\/files\/2026\/04\/10114916\/Blog-LinkedIn.png\" \/>\n\t<meta property=\"og:image:width\" content=\"1200\" \/>\n\t<meta property=\"og:image:height\" content=\"627\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"author\" content=\"Max Liu\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:image\" content=\"https:\/\/static.pingcap.com\/files\/2026\/04\/10114939\/Blog-Feature-Banner.png\" \/>\n<meta name=\"twitter:creator\" content=\"@PingCAP\" \/>\n<meta name=\"twitter:site\" content=\"@PingCAP\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Max Liu\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"6\ubd84\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/thenewstack.io\/tidb-x-open-source-database\/#article\",\"isPartOf\":{\"@id\":\"https:\/\/www.pingcap.com\/blog\/s3-new-network-cloud-object-storage-database-architecture\/\"},\"author\":{\"name\":\"Max Liu\",\"@id\":\"https:\/\/www.pingcap.com\/#\/schema\/person\/7aeccb8936b7f908f923bdd0a6b44be8\"},\"headline\":\"S3 is the New Network: Rethinking Data Architecture for AI Agents\",\"datePublished\":\"2026-04-09T18:52:49+00:00\",\"dateModified\":\"2026-04-10T18:55:01+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/www.pingcap.com\/blog\/s3-new-network-cloud-object-storage-database-architecture\/\"},\"wordCount\":1283,\"publisher\":{\"@id\":\"https:\/\/www.pingcap.com\/#organization\"},\"image\":{\"@id\":\"https:\/\/thenewstack.io\/tidb-x-open-source-database\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/static.pingcap.com\/files\/2026\/04\/10114857\/Blog-Feature.png\",\"keywords\":[\"AI Agents\",\"Distributed SQL\",\"Object Storage\",\"Scalability\",\"TiDB X\"],\"articleSection\":[\"Thought Leadership\"],\"inLanguage\":\"ko-KR\"},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/www.pingcap.com\/blog\/s3-new-network-cloud-object-storage-database-architecture\/\",\"url\":\"https:\/\/thenewstack.io\/tidb-x-open-source-database\/\",\"name\":\"Cloud Object Storage Architecture: S3 Is the New Network\",\"isPartOf\":{\"@id\":\"https:\/\/www.pingcap.com\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/thenewstack.io\/tidb-x-open-source-database\/#primaryimage\"},\"image\":{\"@id\":\"https:\/\/thenewstack.io\/tidb-x-open-source-database\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/static.pingcap.com\/files\/2026\/04\/10114857\/Blog-Feature.png\",\"datePublished\":\"2026-04-09T18:52:49+00:00\",\"dateModified\":\"2026-04-10T18:55:01+00:00\",\"description\":\"Explore the cloud object storage architecture reshaping databases and how TiDB X uses S3 for elastic scale, fast recovery, and AI workloads.\",\"breadcrumb\":{\"@id\":\"https:\/\/thenewstack.io\/tidb-x-open-source-database\/#breadcrumb\"},\"inLanguage\":\"ko-KR\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/thenewstack.io\/tidb-x-open-source-database\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"ko-KR\",\"@id\":\"https:\/\/thenewstack.io\/tidb-x-open-source-database\/#primaryimage\",\"url\":\"https:\/\/static.pingcap.com\/files\/2026\/04\/10114857\/Blog-Feature.png\",\"contentUrl\":\"https:\/\/static.pingcap.com\/files\/2026\/04\/10114857\/Blog-Feature.png\",\"width\":1800,\"height\":600},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/thenewstack.io\/tidb-x-open-source-database\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\/\/www.pingcap.com\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"S3 is the New Network: Rethinking Data Architecture for AI Agents\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.pingcap.com\/#website\",\"url\":\"https:\/\/www.pingcap.com\/\",\"name\":\"TiDB\",\"description\":\"TiDB | SQL at Scale\",\"publisher\":{\"@id\":\"https:\/\/www.pingcap.com\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/www.pingcap.com\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"ko-KR\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/www.pingcap.com\/#organization\",\"name\":\"PingCAP\",\"url\":\"https:\/\/www.pingcap.com\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"ko-KR\",\"@id\":\"https:\/\/www.pingcap.com\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/static.pingcap.com\/files\/2021\/11\/pingcap-logo.png\",\"contentUrl\":\"https:\/\/static.pingcap.com\/files\/2021\/11\/pingcap-logo.png\",\"width\":811,\"height\":232,\"caption\":\"PingCAP\"},\"image\":{\"@id\":\"https:\/\/www.pingcap.com\/#\/schema\/logo\/image\/\"},\"sameAs\":[\"https:\/\/facebook.com\/pingcap2015\",\"https:\/\/x.com\/PingCAP\",\"https:\/\/linkedin.com\/company\/pingcap\",\"https:\/\/youtube.com\/channel\/UCuq4puT32DzHKT5rU1IZpIA\"]},{\"@type\":\"Person\",\"@id\":\"https:\/\/www.pingcap.com\/#\/schema\/person\/7aeccb8936b7f908f923bdd0a6b44be8\",\"name\":\"Max Liu\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"ko-KR\",\"@id\":\"https:\/\/www.pingcap.com\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/static.pingcap.com\/files\/2022\/10\/17234942\/avatar.jpg\",\"contentUrl\":\"https:\/\/static.pingcap.com\/files\/2022\/10\/17234942\/avatar.jpg\",\"caption\":\"Max Liu\"},\"description\":\"Co-Founder &amp; CEO\",\"sameAs\":[\"https:\/\/www.linkedin.com\/in\/max-liu-0b2390147\/\"],\"url\":\"https:\/\/www.pingcap.com\/ko\/blog\/author\/max-liu\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Cloud Object Storage Architecture: S3 Is the New Network","description":"Explore the cloud object storage architecture reshaping databases and how TiDB X uses S3 for elastic scale, fast recovery, and AI workloads.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/thenewstack.io\/tidb-x-open-source-database\/","og_locale":"ko_KR","og_type":"article","og_title":"Cloud Object Storage Architecture: S3 Is the New Network","og_description":"Explore the cloud object storage architecture reshaping databases and how TiDB X uses S3 for elastic scale, fast recovery, and AI workloads.","og_url":"https:\/\/thenewstack.io\/tidb-x-open-source-database\/","og_site_name":"TiDB","article_publisher":"https:\/\/facebook.com\/pingcap2015","article_published_time":"2026-04-09T18:52:49+00:00","article_modified_time":"2026-04-10T18:55:01+00:00","og_image":[{"width":1200,"height":627,"url":"https:\/\/static.pingcap.com\/files\/2026\/04\/10114916\/Blog-LinkedIn.png","type":"image\/png"}],"author":"Max Liu","twitter_card":"summary_large_image","twitter_image":"https:\/\/static.pingcap.com\/files\/2026\/04\/10114939\/Blog-Feature-Banner.png","twitter_creator":"@PingCAP","twitter_site":"@PingCAP","twitter_misc":{"Written by":"Max Liu","Est. reading time":"6\ubd84"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/thenewstack.io\/tidb-x-open-source-database\/#article","isPartOf":{"@id":"https:\/\/www.pingcap.com\/blog\/s3-new-network-cloud-object-storage-database-architecture\/"},"author":{"name":"Max Liu","@id":"https:\/\/www.pingcap.com\/#\/schema\/person\/7aeccb8936b7f908f923bdd0a6b44be8"},"headline":"S3 is the New Network: Rethinking Data Architecture for AI Agents","datePublished":"2026-04-09T18:52:49+00:00","dateModified":"2026-04-10T18:55:01+00:00","mainEntityOfPage":{"@id":"https:\/\/www.pingcap.com\/blog\/s3-new-network-cloud-object-storage-database-architecture\/"},"wordCount":1283,"publisher":{"@id":"https:\/\/www.pingcap.com\/#organization"},"image":{"@id":"https:\/\/thenewstack.io\/tidb-x-open-source-database\/#primaryimage"},"thumbnailUrl":"https:\/\/static.pingcap.com\/files\/2026\/04\/10114857\/Blog-Feature.png","keywords":["AI Agents","Distributed SQL","Object Storage","Scalability","TiDB X"],"articleSection":["Thought Leadership"],"inLanguage":"ko-KR"},{"@type":"WebPage","@id":"https:\/\/www.pingcap.com\/blog\/s3-new-network-cloud-object-storage-database-architecture\/","url":"https:\/\/thenewstack.io\/tidb-x-open-source-database\/","name":"Cloud Object Storage Architecture: S3 Is the New Network","isPartOf":{"@id":"https:\/\/www.pingcap.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/thenewstack.io\/tidb-x-open-source-database\/#primaryimage"},"image":{"@id":"https:\/\/thenewstack.io\/tidb-x-open-source-database\/#primaryimage"},"thumbnailUrl":"https:\/\/static.pingcap.com\/files\/2026\/04\/10114857\/Blog-Feature.png","datePublished":"2026-04-09T18:52:49+00:00","dateModified":"2026-04-10T18:55:01+00:00","description":"Explore the cloud object storage architecture reshaping databases and how TiDB X uses S3 for elastic scale, fast recovery, and AI workloads.","breadcrumb":{"@id":"https:\/\/thenewstack.io\/tidb-x-open-source-database\/#breadcrumb"},"inLanguage":"ko-KR","potentialAction":[{"@type":"ReadAction","target":["https:\/\/thenewstack.io\/tidb-x-open-source-database\/"]}]},{"@type":"ImageObject","inLanguage":"ko-KR","@id":"https:\/\/thenewstack.io\/tidb-x-open-source-database\/#primaryimage","url":"https:\/\/static.pingcap.com\/files\/2026\/04\/10114857\/Blog-Feature.png","contentUrl":"https:\/\/static.pingcap.com\/files\/2026\/04\/10114857\/Blog-Feature.png","width":1800,"height":600},{"@type":"BreadcrumbList","@id":"https:\/\/thenewstack.io\/tidb-x-open-source-database\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/www.pingcap.com\/"},{"@type":"ListItem","position":2,"name":"S3 is the New Network: Rethinking Data Architecture for AI Agents"}]},{"@type":"WebSite","@id":"https:\/\/www.pingcap.com\/#website","url":"https:\/\/www.pingcap.com\/","name":"\ud2f0DB","description":"TiDB | SQL at Scale","publisher":{"@id":"https:\/\/www.pingcap.com\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.pingcap.com\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"ko-KR"},{"@type":"Organization","@id":"https:\/\/www.pingcap.com\/#organization","name":"PingCAP","url":"https:\/\/www.pingcap.com\/","logo":{"@type":"ImageObject","inLanguage":"ko-KR","@id":"https:\/\/www.pingcap.com\/#\/schema\/logo\/image\/","url":"https:\/\/static.pingcap.com\/files\/2021\/11\/pingcap-logo.png","contentUrl":"https:\/\/static.pingcap.com\/files\/2021\/11\/pingcap-logo.png","width":811,"height":232,"caption":"PingCAP"},"image":{"@id":"https:\/\/www.pingcap.com\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/facebook.com\/pingcap2015","https:\/\/x.com\/PingCAP","https:\/\/linkedin.com\/company\/pingcap","https:\/\/youtube.com\/channel\/UCuq4puT32DzHKT5rU1IZpIA"]},{"@type":"Person","@id":"https:\/\/www.pingcap.com\/#\/schema\/person\/7aeccb8936b7f908f923bdd0a6b44be8","name":"Max Liu","image":{"@type":"ImageObject","inLanguage":"ko-KR","@id":"https:\/\/www.pingcap.com\/#\/schema\/person\/image\/","url":"https:\/\/static.pingcap.com\/files\/2022\/10\/17234942\/avatar.jpg","contentUrl":"https:\/\/static.pingcap.com\/files\/2022\/10\/17234942\/avatar.jpg","caption":"Max Liu"},"description":"Co-Founder &amp; CEO","sameAs":["https:\/\/www.linkedin.com\/in\/max-liu-0b2390147\/"],"url":"https:\/\/www.pingcap.com\/ko\/blog\/author\/max-liu\/"}]}},"grav_blocks":false,"card_markup":"<a class=\"card-resource bg-white\" href=\"https:\/\/www.pingcap.com\/ko\/blog\/s3-new-network-cloud-object-storage-database-architecture\/\"><div class=\"card-resource__image-container\"><img class=\"card-resource__image\" alt=\"Blog - Feature\" src=\"https:\/\/static.pingcap.com\/files\/2026\/04\/10114857\/Blog-Feature.png\" loading=\"lazy\" width=1800 height=600 \/><\/div><div class=\"card-resource__content-container\"><div class=\"card-resource__content-head\"><div class=\"card-resource__category\">Thought Leadership<\/div><\/div><h5 class=\"card-resource__title\">S3 is the New Network: Rethinking Data Architecture for AI Agents<\/h5><\/div><\/a>","_links":{"self":[{"href":"https:\/\/www.pingcap.com\/ko\/wp-json\/wp\/v2\/posts\/32939","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.pingcap.com\/ko\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.pingcap.com\/ko\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.pingcap.com\/ko\/wp-json\/wp\/v2\/users\/50"}],"replies":[{"embeddable":true,"href":"https:\/\/www.pingcap.com\/ko\/wp-json\/wp\/v2\/comments?post=32939"}],"version-history":[{"count":9,"href":"https:\/\/www.pingcap.com\/ko\/wp-json\/wp\/v2\/posts\/32939\/revisions"}],"predecessor-version":[{"id":32955,"href":"https:\/\/www.pingcap.com\/ko\/wp-json\/wp\/v2\/posts\/32939\/revisions\/32955"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.pingcap.com\/ko\/wp-json\/wp\/v2\/media\/32948"}],"wp:attachment":[{"href":"https:\/\/www.pingcap.com\/ko\/wp-json\/wp\/v2\/media?parent=32939"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.pingcap.com\/ko\/wp-json\/wp\/v2\/categories?post=32939"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.pingcap.com\/ko\/wp-json\/wp\/v2\/tags?post=32939"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}