{"@context":"https://schema.org","@type":"NewsArticle","generatedAt":"2026-07-28T06:20:51.496Z","headline":"AlloyDB 推出列式引擎加速 HNSW，pgvector 向量搜索 QPS 提升最高 4.9 倍","description":"Google Cloud 的 AlloyDB 在预览版中推出列式引擎加速的 HNSW 索引，使 pgvector 向量搜索的每秒查询数（QPS）相比标准 PostgreSQL HNSW 提升最高 4.9 倍，在固定 QPS 下召回率可提升 0.163。该加速通过将索引固定在列式引擎内存中、使用向量化内存布局并绕过标准 PostgreSQL 缓冲管理器开销实现，无需修改应用即可获得性能提升。","url":"https://www.aioga.com/news/cms3dqo6s0axdro3fg3b4mfc0/","mainEntityOfPage":"https://www.aioga.com/news/cms3dqo6s0axdro3fg3b4mfc0/","datePublished":"2026-07-21T16:00:00.000Z","dateModified":"2026-07-21T16:00:00.000Z","inLanguage":"zh-CN","publisher":{"@type":"NewsMediaOrganization","name":"Aioga","url":"https://www.aioga.com"},"citation":["https://cloud.google.com/blog/products/databases/supercharge-pgvector-4x-faster-hnsw-with-alloydb","https://aihot.virxact.com/items/cms3dqo6s0axdro3fg3b4mfc0"],"canonicalUrl":"https://www.aioga.com/news/cms3dqo6s0axdro3fg3b4mfc0/","directAnswer":{"@type":"Answer","text":"Google Cloud 表示，AlloyDB 预览版新增列式引擎加速的 HNSW 索引。材料称，pgvector 向量搜索 QPS 相比标准 PostgreSQL HNSW 最高提升 4.9 倍，固定 QPS 下召回率可提升 0.163，且无需修改应用。","url":"https://www.aioga.com/news/cms3dqo6s0axdro3fg3b4mfc0/","dateCreated":"2026-07-21T16:00:00.000Z","author":{"@type":"Organization","@id":"https://www.aioga.com/authors/aioga-editorial/#editorial-team","name":"Aioga Editorial Team","url":"https://www.aioga.com/authors/aioga-editorial/"}},"evidence":[{"@type":"CreativeWork","name":"cloud.google.com source article","url":"https://cloud.google.com/blog/products/databases/supercharge-pgvector-4x-faster-hnsw-with-alloydb","datePublished":"2026-07-21T16:00:00.000Z","provider":{"@type":"Organization","name":"cloud.google.com","url":"https://cloud.google.com/blog/products/databases/supercharge-pgvector-4x-faster-hnsw-with-alloydb"}},{"@type":"CreativeWork","name":"AIHot archive record","url":"https://aihot.virxact.com/items/cms3dqo6s0axdro3fg3b4mfc0","datePublished":"2026-07-21T16:00:00.000Z","provider":{"@type":"Organization","name":"AIHot","url":"https://aihot.virxact.com/items/cms3dqo6s0axdro3fg3b4mfc0"}}],"aggregationSource":"Google Cloud：Databases（RSS）","originalPublisher":{"name":"cloud.google.com","url":"https://cloud.google.com/blog/products/databases/supercharge-pgvector-4x-faster-hnsw-with-alloydb"},"article":{"id":"cms3dqo6s0axdro3fg3b4mfc0","slug":"cms3dqo6s0axdro3fg3b4mfc0","url":"https://www.aioga.com/news/cms3dqo6s0axdro3fg3b4mfc0/","title":"AlloyDB 推出列式引擎加速 HNSW，pgvector 向量搜索 QPS 提升最高 4.9 倍","title_en":"Supercharging pgvector： 4x faster HNSW vector search with AlloyDB","summary":"Google Cloud 的 AlloyDB 在预览版中推出列式引擎加速的 HNSW 索引，使 pgvector 向量搜索的每秒查询数（QPS）相比标准 PostgreSQL HNSW 提升最高 4.9 倍，在固定 QPS 下召回率可提升 0.163。该加速通过将索引固定在列式引擎内存中、使用向量化内存布局并绕过标准 PostgreSQL 缓冲管理器开销实现，无需修改应用即可获得性能提升。","source":"Google Cloud：Databases（RSS）","sourceUrl":"https://cloud.google.com/blog/products/databases/supercharge-pgvector-4x-faster-hnsw-with-alloydb","aiHotUrl":"https://aihot.virxact.com/items/cms3dqo6s0axdro3fg3b4mfc0","publishedAt":"2026-07-21T16:00:00.000Z","category":"产品更新","score":71,"selected":true,"articleBody":["The front door to AI in the workplace","AlloyDB ：https://cloud.google.com/alloydb is a fully managed, PostgreSQL-compatible database service built for your most demanding enterprise workloads. It combines the best of open source PostgreSQL with Google’s advanced technology, offering massive scalability, high availability, and native AI capabilities. It serves as a performant relational store, a unified backend for vector and full text search, and an analytics engine that is up to 100x faster than standard PostgreSQL.","Vector search is the foundation of modern AI and Retrieval Augmented Generation (RAG) applications. For developers using AlloyDB and other PostgreSQL databases, pgvector ：https://github.com/pgvector/pgvector is a widely adopted extension for storing, indexing, and querying vector embeddings, and HNSW (Hierarchical Navigable Small World) is a highly efficient graph-based algorithm designed for approximate nearest neighbor search across multi-layered structures. With columnar engine accelerated HNSW ：https://docs.cloud.google.com/alloydb/docs/ai/accelerate-with-ce in AlloyDB (now in preview), you can achieve up to 4x higher queries per second (QPS) for vector search compared to standard PostgreSQL HNSW.","Enterprise AI applications face a constant trade-off between speed and accuracy. When searching through millions or billions of vectors, maximizing Queries per Second (QPS) without sacrificing search quality (recall) is critical for scaling production workloads. The PostgreSQL pgvector extension offers HNSW as one of the indexes that can speed up Approximate Nearest Neighbor (ANN) searches. Let’s dive deep into how AlloyDB solves the speed vs. accuracy trade-off.","Note: While this post focuses on HNSW performance, it’s worth noting that HNSW is just one part of AlloyDB’s advanced vector toolkit. AlloyDB also features ScaNN ：https://docs.cloud.google.com/alloydb/docs/ai/create-scann-index —a cutting-edge index backed by over 14 years of Google Research—giving you the flexibility to choose the perfect index for your workload. Additionally, for use cases demanding absolute precision, standard k-nearest neighbor (KNN) search is always available for 100% recall. Check out our Choose a Vector Index Guide ：https://docs.cloud.google.com/alloydb/docs/ai/choose-index-strategy to see how they stack up.","AlloyDB offers a free trial instance ：https://cloud.google.com/alloydb/docs/free-trial-cluster for you to try AlloyDB with your own workload. In addition, new Google Cloud customers get $300 in free credits.","The AlloyDB columnar engine ：https://docs.cloud.google.com/alloydb/docs/columnar-engine/about is a built-in, in-memory cache that automatically stores frequently queried data in a specialized, scan-optimized columnar format. It allows AlloyDB to handle heavy analytical queries up to 100x faster than standard PostgreSQL. Additionally, it accelerates ANN searches by storing the index in memory, using a vectorized memory layout for fast traversals, and bypassing standard PostgreSQL buffer manager overhead.","To understand the real-world performance characteristics of columnar engine Accelerated HNSW, we plotted standard QPS vs Recall curves for the GloVe 100 Angular dataset by searching more than 1M records with a limit of 100.","Running this benchmark script ：https://colab.research.google.com/github/GoogleCloudPlatform/python-docs-samples/blob/main/alloydb/notebooks/columnar_engine_accelerated_hnsw_vector_search_benchmark.ipynb yields the following visualization:","Note: These measurements were taken on an AlloyDB C4A 16vCPU machine. Due to the inherent randomness in HNSW graph building, results may slightly vary across runs.","The data reveals two transformative benefits:","It is important to note that the baseline (blue line) already represents the index being fully cached in the PostgreSQL shared buffer cache. The performance gains shown here are not the result of moving data from disk to RAM, but rather the result of a more efficient memory architecture.","In standard PostgreSQL architectures, index operations utilize the shared buffer cache. Even when data is fully in-memory, the database still incurs significant overhead from the buffer manager, which must handle operations such as page pinning and unpinning, lock acquisition, buffer table lookups, and Least Recently Used (LRU) management.","AlloyDB's columnar engine is a built-in, in-memory cache that stores data in a specialized, scan-optimized format.","With this release, AlloyDB can use columnar engine accelerated HNSW to:","For enterprise-scale applications, this isn't just about a faster database—it's about cost and quality:","Reduced infrastructure costs: Achieve the same performance with significantly lower compute resources.","Better AI accuracy: Reach higher recall and quality at speeds that were previously only possible for \"draft\" (high-speed, lower-accuracy results) quality search.","No application changes required: Because this is built into AlloyDB, you get these gains using the same standard pgvector SQL syntax.","Note that the columnar engine does utilize memory, but it is highly compressed and meticulously managed. Because the engine stores vector data in an efficient columnar format, the memory footprint is minimal compared to the massive performance gains—making it a highly favorable trade-off for enterprise workloads.","To try out columnar engine accelerated HNSW ：https://docs.cloud.google.com/alloydb/docs/ai/accelerate-with-ce in AlloyDB, follow these steps:","1. Enable the columnar engine ：https://docs.cloud.google.com/alloydb/docs/ai/accelerate-with-ce and index caching","Ensure that both google_columnar_engine.enabled and google_columnar_engine.enable_index_caching flags are set to on for your AlloyDB instance.","2. Add the HNSW Index to columnar engine","Once your HNSW index is created via pgvector , execute the following SQL command to cache it in the columnar engine:","New to AlloyDB? Discover AlloyDB with a 30-day free trial ：https://docs.cloud.google.com/alloydb/docs/free-trial-cluster .","Google Colab Notebook ：https://colab.research.google.com/github/GoogleCloudPlatform/python-docs-samples/blob/main/alloydb/notebooks/columnar_engine_accelerated_hnsw_vector_search_benchmark.ipynb : An end-to-end Python script to ingest the GloVe dataset, create indexes, and plot Recall vs QPS curves.","Is HNSW the right vector index choice for your use case? Check our ‘ Choose a vector index in AlloyDB AI ：https://docs.cloud.google.com/alloydb/docs/ai/choose-index-strategy ’ guide.","By Sirish Chandrasekaran • 4-minute read"],"articleImages":[{"sourceUrl":"https://storage.googleapis.com/gweb-cloudblog-publish/images/1_r57mjyN.max-1200x1200.png","alt":"https://storage.googleapis.com/gweb-cloudblog-publish/images/1_r57mjyN.max-1200x1200.png","afterParagraph":8,"url":"/media/articles/cms3dqo6s0axdro3fg3b4mfc0/2c894d4f2a27f35a.png"},{"sourceUrl":"https://storage.googleapis.com/gweb-cloudblog-publish/images/1_TdmG649.max-700x700.png","alt":"https://storage.googleapis.com/gweb-cloudblog-publish/images/1_TdmG649.max-700x700.png","afterParagraph":27,"url":"/media/articles/cms3dqo6s0axdro3fg3b4mfc0/d2fb20ee224fcc86.png"},{"sourceUrl":"https://storage.googleapis.com/gweb-cloudblog-publish/images/10_-_Databases.max-700x700.jpg","alt":"https://storage.googleapis.com/gweb-cloudblog-publish/images/10_-_Databases.max-700x700.jpg","afterParagraph":28,"url":"/media/articles/cms3dqo6s0axdro3fg3b4mfc0/6ab573326db35ef3.jpg"},{"sourceUrl":"https://www.gstatic.com/cgc/super_cloud_gradient.png","alt":"Cloud logo","afterParagraph":28,"url":"/media/articles/cms3dqo6s0axdro3fg3b4mfc0/9587dad254ef77a2.png"}],"mediaStatus":"ok","articleBodyZh":["工作场所人工智能的前门","AlloyDB：https://cloud.google.com/alloydb 是一个完全托管的、与 PostgreSQL 兼容的数据库服务，专为最苛刻的企业工作负载而构建。它结合了开源 PostgreSQL 的优点与谷歌的先进技术，提供极大的可扩展性、高可用性和原生 AI 功能。它既可以作为高性能的关系型存储，也可以作为向量和全文搜索的统一后端，同时作为分析引擎，其速度比标准 PostgreSQL 快 100 倍。","向量搜索是现代 AI 和增强检索生成（RAG）应用的基础。对于使用 AlloyDB 和其他 PostgreSQL 数据库的开发者来说，pgvector：https://github.com/pgvector/pgvector 是一个被广泛采用的扩展，用于存储、索引和查询向量嵌入，而 HNSW（分层可导航小世界）是一种高效的基于图的算法，设计用于多层结构的近似最近邻搜索。通过 AlloyDB 中的列式引擎加速 HNSW：https://docs.cloud.google.com/alloydb/docs/ai/accelerate-with-ce（现正处于预览阶段），在向量搜索中可实现比标准 PostgreSQL HNSW 高出 4 倍的每秒查询量（QPS）。","企业 AI 应用面临速度与准确性之间的持续权衡。在搜索数百万或数十亿个向量时，最大化每秒查询量（QPS）而不牺牲搜索质量（召回率），对于扩大生产工作负载至关重要。PostgreSQL 的 pgvector 扩展提供 HNSW 作为可以加速近似最近邻（ANN）搜索的索引之一。让我们深入探讨 AlloyDB 如何解决速度与准确性的权衡问题。","注意：虽然本文关注的是 HNSW 性能，但值得注意的是 HNSW 只是 AlloyDB 高级向量工具包的一部分。AlloyDB 还提供 ScaNN：https://docs.cloud.google.com/alloydb/docs/ai/create-scann-index ——由 Google Research 超过 14 年的研究支持的前沿索引——为您提供选择最适合工作负载的索引的灵活性。此外，对于需要绝对精度的用例，标准的 k 最近邻（KNN）搜索始终可用于 100% 召回率。请查看我们的“选择向量索引指南”：https://docs.cloud.google.com/alloydb/docs/ai/choose-index-strategy，了解它们的比较情况。","AlloyDB 提供免费试用实例：https://cloud.google.com/alloydb/docs/free-trial-cluster，让您可以使用自己的工作负载体验 AlloyDB。此外，新 Google Cloud 用户可获得 300 美元的免费额度。","AlloyDB 列式引擎：https://docs.cloud.google.com/alloydb/docs/columnar-engine/about 是一个内置的内存缓存，自动以专门的、扫描优化的列式格式存储经常查询的数据。它使 AlloyDB 能够处理高达比标准 PostgreSQL 快 100 倍的重分析查询。此外，它通过将索引存储在内存中、使用向量化内存布局进行快速遍历，并绕过标准 PostgreSQL 缓冲区管理器开销，从而加速 ANN 搜索。","为了了解列式引擎加速 HNSW 的现实性能特征，我们绘制了 GloVe 100 Angular 数据集的标准 QPS 对召回率曲线，通过在超过 100 万条记录中以 100 条限制进行搜索。","运行此基准测试脚本：https://colab.research.google.com/github/GoogleCloudPlatform/python-docs-samples/blob/main/alloydb/notebooks/columnar_engine_accelerated_hnsw_vector_search_benchmark.ipynb 得到以下可视化结果：","注意：这些测量是在 AlloyDB C4A 16vCPU 机器上进行的。由于 HNSW 图构建的固有随机性，结果在不同运行间可能略有差异。","数据揭示了两个变革性益处：","需要注意的是，基线（蓝色线）已经表示索引完全缓存在 PostgreSQL 共享缓冲缓存中。这里显示的性能提升并不是将数据从磁盘移动到内存的结果，而是更高效的内存架构带来的结果。","在标准的 PostgreSQL 架构中，索引操作会使用共享缓冲缓存。即使数据完全驻留在内存中，数据库仍然会产生缓冲管理器的显著开销，缓冲管理器需要处理诸如页面固定与解固定、锁获取、缓冲表查找以及最近最少使用（LRU）管理等操作。","AlloyDB 的列存引擎是内置的内存缓存，以专门的、优化扫描的格式存储数据。","在此版本中，AlloyDB 可以使用列存引擎加速的 HNSW 来：","对于企业级应用，这不仅仅是一个更快的数据库问题——而是成本和质量问题：","降低基础设施成本：以显著更低的计算资源实现相同性能。","更佳的 AI 准确性：以以前仅可用于“草稿”（高速、低精度结果）质量搜索的速度，实现更高的召回率和质量。","无需更改应用程序：因为这是内置于 AlloyDB 中的，因此可以使用相同的标准 pgvector SQL 语法获得这些提升。","注意列存引擎确实会使用内存，但它经过高度压缩并经过精心管理。由于引擎以高效的列存格式存储向量数据，相对于巨大的性能提升，内存占用极小——这对于企业工作负载来说是一个非常有利的权衡。","要在 AlloyDB 中试用列存引擎加速的 HNSW，请访问： https://docs.cloud.google.com/alloydb/docs/ai/accelerate-with-ce，并按照以下步骤操作：","1. 启用列存引擎： https://docs.cloud.google.com/alloydb/docs/ai/accelerate-with-ce 并启用索引缓存","确保在 AlloyDB 实例中将 google_columnar_engine.enabled 和 google_columnar_engine.enable_index_caching 标志都设置为开启状态。","2. 将 HNSW 索引添加到列存引擎","一旦通过 pgvector 创建了 HNSW 索引，执行以下 SQL 命令将其缓存在列存引擎中：","初次使用 AlloyDB？通过 30 天免费试用了解 AlloyDB：https://docs.cloud.google.com/alloydb/docs/free-trial-cluster。","Google Colab 笔记本：https://colab.research.google.com/github/GoogleCloudPlatform/python-docs-samples/blob/main/alloydb/notebooks/columnar_engine_accelerated_hnsw_vector_search_benchmark.ipynb：一个端到端的 Python 脚本，用于导入 GloVe 数据集、创建索引，并绘制召回率与 QPS 曲线。","HNSW 是您用例中合适的向量索引选择吗？查看我们的《AlloyDB AI 中选择向量索引》指南：https://docs.cloud.google.com/alloydb/docs/ai/choose-index-strategy。","作者：Sirish Chandrasekaran • 阅读时间：4 分钟"],"translationStatus":"translated","bodyOrigin":"source-page","editorial":{"summary":"Google Cloud 表示，AlloyDB 预览版新增列式引擎加速的 HNSW 索引。材料称，pgvector 向量搜索 QPS 相比标准 PostgreSQL HNSW 最高提升 4.9 倍，固定 QPS 下召回率可提升 0.163，且无需修改应用。","background":"HNSW 是 pgvector 支持的近似最近邻搜索索引，常用于向量检索和 RAG 应用。AlloyDB 将索引固定在列式引擎内存中，采用向量化内存布局，并绕过标准 PostgreSQL 缓冲管理器开销。","viewpoint":"Aioga 判断，这项更新的核心价值在于尝试缓解向量搜索中速度与召回率之间的权衡。不过材料仅介绍预览版及特定测试结果，实际收益仍可能取决于数据集、查询负载和部署配置。","implications":"对已使用 AlloyDB 与 pgvector 的团队而言，列式引擎加速 HNSW 可能降低提升检索吞吐的应用改造成本。材料同时提到 AlloyDB 还提供 ScaNN 和标准 KNN，索引选择仍需结合精度与性能需求。","nextStep":"值得关注的是该能力从预览版走向正式可用后的性能表现、适用限制与成本信息。评估团队可依据自身向量数据和查询模式进行测试，并对比 HNSW、ScaNN 及标准 KNN 的结果。","evidenceRefs":["title","summary","articleBody","source"],"status":"published","aiGenerated":true,"autoApproved":true,"generatedBy":"aioga-editorial:gpt-5.6-sol","reviewedBy":"aioga-editorial-review:gpt-5.6-sol","generatedAt":"2026-07-27T15:53:58.203Z","sourceHash":"3f5101cfe639b9f1","review":{"approved":true,"groundedness":92,"clarity":90,"duplicationRisk":24,"blockingIssues":[],"notes":["来源材料内部对 QPS 提升幅度存在“最高 4.9 倍”与正文摘录“最高 4 倍”的表述差异；候选内容采用了标题和摘要中的 4.9 倍，仍有来源依据，但可注明具体测试条件以避免被理解为普遍性能保证。","“实际收益仍可能取决于数据集、查询负载和部署配置”属于合理且已明确标示为判断的审慎说明，不构成无来源事实断言。","“可能降低提升检索吞吐的应用改造成本”由“无需修改应用即可获得性能提升”合理推导，并使用了不确定性措辞。"]},"validation":{"passed":true,"mode":"ai-auto","revisions":0,"checks":["schema","length","source-attribution","low-source-overlap","no-html","independent-ai-review"]}},"tags":["产品更新","Google Cloud：Databases（RSS）"],"translations":{"zh-CN":{"title":"AlloyDB 推出列式引擎加速 HNSW，pgvector 向量搜索 QPS 提升最高 4.9 倍","summary":"Google Cloud 的 AlloyDB 在预览版中推出列式引擎加速的 HNSW 索引，使 pgvector 向量搜索的每秒查询数（QPS）相比标准 PostgreSQL HNSW 提升最高 4.9 倍，在固定 QPS 下召回率可提升 0.163。该加速通过将索引固定在列式引擎内存中、使用向量化内存布局并绕过标准 PostgreSQL 缓冲管理器开销实现，无需修改应用即可获得性能提升。","category":"产品更新","source":"cloud.google.com","aggregationSource":"Google Cloud：Databases（RSS）","pageTitle":"AlloyDB 推出列式引擎加速 HNSW，pgvector 向量搜索 QPS 提升最高 4.9 倍 - Aioga AI资讯","description":"Google Cloud 的 AlloyDB 在预览版中推出列式引擎加速的 HNSW 索引，使 pgvector 向量搜索的每秒查询数（QPS）相比标准 PostgreSQL HNSW 提升最高 4.9 倍，在固定 QPS 下召回率可提升 0.163。该加速通过将索引固定在列式引擎内存中、使用向量化内存布局并绕过标准 PostgreSQL 缓冲管理器开销实现...","url":"https://www.aioga.com/news/cms3dqo6s0axdro3fg3b4mfc0/"},"en":{"title":"AlloyDB launches columnar engine to accelerate HNSW, pgvector vector search QPS increased by up to 4.9 times","summary":"Google Cloud's AlloyDB has launched a columnar engine-accelerated HNSW index in the preview version, enabling pgvector vector search to achieve up to 4.9 times higher queries per second (QPS) compared to standard PostgreSQL HNSW, with recall increasing by 0.163 at fixed QPS. This acceleration is achieved by keeping the index in the columnar engine memory, using a vectorized memory layout, and bypassing the overhead of the standard PostgreSQL buffer manager, allowing performance improvements without modifying the application.","category":"Products","source":"Google Cloud：Databases（RSS）","aggregationSource":"Google Cloud：Databases（RSS）","pageTitle":"AlloyDB launches columnar engine to accelerate HNSW, pgvector vector search QPS increased by up to 4.9 times - Aioga AI News","description":"Google Cloud's AlloyDB has launched a columnar engine-accelerated HNSW index in the preview version, enabling pgvector vector search to achieve up to 4.9 times higher queries per s...","url":"https://www.aioga.com/en/news/cms3dqo6s0axdro3fg3b4mfc0/","contentTranslated":true,"sourceHash":"30fec604009e3a2d","translatedAt":"2026-07-27T15:46:04.970Z"},"ja":{"title":"AlloyDB、列指向エンジンで HNSW を加速、pgvector ベクトル検索 QPS 最大 4.9 倍向上","summary":"Google Cloud の AlloyDB はプレビュー版で列指向エンジンを加速する HNSW インデックスを導入し、pgvector ベクトル検索の毎秒クエリ数（QPS）が標準の PostgreSQL HNSW に比べ最大 4.9 倍に向上し、固定 QPS 下での再現率が 0.163 改善されます。この加速は、インデックスを列指向エンジンのメモリ内に固定し、ベクトル化されたメモリレイアウトを使用し、標準の PostgreSQL バッファ管理オーバーヘッドを回避することで実現され、アプリケーションを変更することなく性能向上を得ることができます。","category":"製品更新","source":"Google Cloud：Databases（RSS）","aggregationSource":"Google Cloud：Databases（RSS）","pageTitle":"AlloyDB、列指向エンジンで HNSW を加速、pgvector ベクトル検索 QPS 最大 4.9 倍向上 - Aioga AIニュース","description":"Google Cloud の AlloyDB はプレビュー版で列指向エンジンを加速する HNSW インデックスを導入し、pgvector ベクトル検索の毎秒クエリ数（QPS）が標準の PostgreSQL HNSW に比べ最大 4.9 倍に向上し、固定 QPS 下での再現率が 0.163 改善されます。この加速は、インデックスを列指向エンジンのメモリ内に固定...","url":"https://www.aioga.com/ja/news/cms3dqo6s0axdro3fg3b4mfc0/","contentTranslated":true,"sourceHash":"30fec604009e3a2d","translatedAt":"2026-07-27T15:46:13.988Z"},"ko":{"title":"AlloyDB는 컬럼형 엔진으로 HNSW를 가속화하고, pgvector 벡터 검색 QPS를 최대 4.9배 향상시켰습니다","summary":"Google Cloud의 AlloyDB는 미리보기 버전에서 열 지향 엔진으로 가속된 HNSW 인덱스를 출시하여, pgvector 벡터 검색의 초당 쿼리 수(QPS)가 표준 PostgreSQL HNSW에 비해 최대 4.9배 향상되고, 고정 QPS에서 재현율이 0.163 향상될 수 있습니다. 이러한 가속은 인덱스를 열 지향 엔진 메모리에 고정하고, 벡터화된 메모리 레이아웃을 사용하며, 표준 PostgreSQL 버퍼 관리자 오버헤드를 우회함으로써 구현되며, 애플리케이션을 수정하지 않고도 성능 향상을 얻을 수 있습니다.","category":"제품 업데이트","source":"Google Cloud：Databases（RSS）","aggregationSource":"Google Cloud：Databases（RSS）","pageTitle":"AlloyDB는 컬럼형 엔진으로 HNSW를 가속화하고, pgvector 벡터 검색 QPS를 최대 4.9배 향상시켰습니다 - Aioga AI 뉴스","description":"Google Cloud의 AlloyDB는 미리보기 버전에서 열 지향 엔진으로 가속된 HNSW 인덱스를 출시하여, pgvector 벡터 검색의 초당 쿼리 수(QPS)가 표준 PostgreSQL HNSW에 비해 최대 4.9배 향상되고, 고정 QPS에서 재현율이 0.163 향상될 수 있습니다. 이러한 가속은 인덱스를 열 지향...","url":"https://www.aioga.com/ko/news/cms3dqo6s0axdro3fg3b4mfc0/","contentTranslated":true,"sourceHash":"30fec604009e3a2d","translatedAt":"2026-07-27T15:47:12.071Z"},"es":{"title":"AlloyDB lanza motor columnar para acelerar HNSW, la búsqueda de vectores pgvector aumenta el QPS hasta 4,9 veces","summary":"AlloyDB de Google Cloud lanza en versión preliminar un índice HNSW acelerado por motor columnar, lo que permite que la búsqueda de vectores pgvector aumente hasta 4.9 veces las consultas por segundo (QPS) en comparación con el HNSW estándar de PostgreSQL, y mejore la tasa de recuperación en 0.163 con una QPS fija. Esta aceleración se logra fijando el índice en la memoria del motor columnar, utilizando un diseño de memoria vectorizado y omitiendo la sobrecarga del gestor de búfer estándar de PostgreSQL, sin necesidad de modificar la aplicación para obtener la mejora de rendimiento.","category":"Productos","source":"Google Cloud：Databases（RSS）","aggregationSource":"Google Cloud：Databases（RSS）","pageTitle":"AlloyDB lanza motor columnar para acelerar HNSW, la búsqueda de vectores pgvector aumenta el QPS hasta 4,9 veces - Aioga Noticias de IA","description":"AlloyDB de Google Cloud lanza en versión preliminar un índice HNSW acelerado por motor columnar, lo que permite que la búsqueda de vectores pgvector aumente hasta 4.9 veces las con...","url":"https://www.aioga.com/es/news/cms3dqo6s0axdro3fg3b4mfc0/","contentTranslated":true,"sourceHash":"30fec604009e3a2d","translatedAt":"2026-07-27T15:47:00.121Z"},"fr":{"title":"AlloyDB lance un moteur en colonnes pour accélérer HNSW, la recherche vectorielle pgvector augmente le QPS jusqu'à 4,9 fois","summary":"AlloyDB de Google Cloud a lancé en version préliminaire un index HNSW accéléré par moteur en colonnes, permettant à la recherche vectorielle pgvector d’atteindre jusqu’à 4,9 fois le nombre de requêtes par seconde (QPS) par rapport à l’HNSW standard de PostgreSQL, et d’améliorer le taux de rappel de 0,163 à QPS fixe. Cette accélération est réalisée en fixant l’index dans la mémoire du moteur en colonnes, en utilisant une disposition mémoire vectorisée et en contournant les frais généraux du gestionnaire de tampons standard de PostgreSQL, sans nécessiter de modification de l’application pour obtenir l’amélioration des performances.","category":"Produits","source":"Google Cloud：Databases（RSS）","aggregationSource":"Google Cloud：Databases（RSS）","pageTitle":"AlloyDB lance un moteur en colonnes pour accélérer HNSW, la recherche vectorielle pgvector augmente le QPS jusqu'à 4,9 fois - Aioga Actualités IA","description":"AlloyDB de Google Cloud a lancé en version préliminaire un index HNSW accéléré par moteur en colonnes, permettant à la recherche vectorielle pgvector d’atteindre jusqu’à 4,9 fois l...","url":"https://www.aioga.com/fr/news/cms3dqo6s0axdro3fg3b4mfc0/","contentTranslated":true,"sourceHash":"30fec604009e3a2d","translatedAt":"2026-07-27T15:47:59.105Z"},"de":{"title":"AlloyDB führt eine spaltenbasierte Engine zur Beschleunigung von HNSW ein, pgvector Vektor-Suche QPS-Steigerung um bis zu 4,9-fach","summary":"Google Cloud hat AlloyDB in der Vorschauversion mit einer spaltenbasierten Engine-beschleunigten HNSW-Indizierung eingeführt, wodurch die Anfragen pro Sekunde (QPS) der pgvector-Vektorsuche im Vergleich zur standardmäßigen PostgreSQL-HNSW-Indizierung um bis zu 4,9 Mal gesteigert werden können, während bei fester QPS die Rückrufquote um 0,163 verbessert werden kann. Diese Beschleunigung wird erreicht, indem der Index im Speicher der spaltenbasierten Engine fixiert wird, ein vektorisierter Speicherlayout verwendet wird und die Standard-PostgreSQL-Pufferverwaltung umgangen wird, ohne dass die Anwendung modifiziert werden muss, um Leistungsverbesserungen zu erzielen.","category":"产品更新","source":"Google Cloud：Databases（RSS）","aggregationSource":"Google Cloud：Databases（RSS）","pageTitle":"AlloyDB führt eine spaltenbasierte Engine zur Beschleunigung von HNSW ein, pgvector Vektor-Suche QPS-Steigerung um bis zu 4,9-fach - Aioga KI-News","description":"Google Cloud hat AlloyDB in der Vorschauversion mit einer spaltenbasierten Engine-beschleunigten HNSW-Indizierung eingeführt, wodurch die Anfragen pro Sekunde (QPS) der pgvector-Ve...","url":"https://www.aioga.com/de/news/cms3dqo6s0axdro3fg3b4mfc0/","contentTranslated":true,"sourceHash":"30fec604009e3a2d","translatedAt":"2026-07-27T15:47:57.547Z"},"pt-BR":{"title":"AlloyDB lança motor columnar para acelerar HNSW, pesquisa vetorial pgvector aumenta QPS em até 4,9 vezes","summary":"O AlloyDB do Google Cloud lançou na versão de visualização um índice HNSW acelerado por mecanismo colunar, permitindo que a busca de vetores pgvector aumente o número de consultas por segundo (QPS) em até 4,9 vezes em comparação com o HNSW padrão do PostgreSQL, e a taxa de recall pode ser aumentada em 0,163 em QPS fixo. Essa aceleração é obtida mantendo o índice na memória do mecanismo colunar, usando um layout de memória vetorizado e contornando o overhead do gerenciador de buffer padrão do PostgreSQL, sem necessidade de modificar a aplicação para obter o aumento de desempenho.","category":"产品更新","source":"Google Cloud：Databases（RSS）","aggregationSource":"Google Cloud：Databases（RSS）","pageTitle":"AlloyDB lança motor columnar para acelerar HNSW, pesquisa vetorial pgvector aumenta QPS em até 4,9 vezes - Aioga Notícias de IA","description":"O AlloyDB do Google Cloud lançou na versão de visualização um índice HNSW acelerado por mecanismo colunar, permitindo que a busca de vetores pgvector aumente o número de consultas...","url":"https://www.aioga.com/pt-BR/news/cms3dqo6s0axdro3fg3b4mfc0/","contentTranslated":true,"sourceHash":"30fec604009e3a2d","translatedAt":"2026-07-27T15:48:40.754Z"},"ru":{"title":"AlloyDB представила столбцовый движок для ускорения HNSW, производительность поиска вектора pgvector увеличена до 4,9 раз","summary":"AlloyDB от Google Cloud в предварительной версии представила ускоренный столбцовый движок с индексом HNSW, благодаря чему количество запросов в секунду (QPS) для поиска вектора pgvector увеличилось до 4,9 раза по сравнению со стандартным PostgreSQL HNSW, а при фиксированном QPS точность извлечения увеличилась на 0,163. Это ускорение достигается за счёт закрепления индекса в памяти столбцового движка, использования векторизованного расположения памяти и обхода стандартного буферного менеджера PostgreSQL, при этом для получения прироста производительности не требуется модифицировать приложения.","category":"产品更新","source":"Google Cloud：Databases（RSS）","aggregationSource":"Google Cloud：Databases（RSS）","pageTitle":"AlloyDB представила столбцовый движок для ускорения HNSW, производительность поиска вектора pgvector увеличена до 4,9 раз - Aioga Новости ИИ","description":"AlloyDB от Google Cloud в предварительной версии представила ускоренный столбцовый движок с индексом HNSW, благодаря чему количество запросов в секунду (QPS) для поиска вектора pgv...","url":"https://www.aioga.com/ru/news/cms3dqo6s0axdro3fg3b4mfc0/","contentTranslated":true,"sourceHash":"30fec604009e3a2d","translatedAt":"2026-07-27T15:48:47.692Z"},"ar":{"title":"أطلقت AlloyDB محرك أعمدة لتسريع HNSW، وزادت سرعة بحث المتجهات pgvector QPS حتى 4.9 مرات","summary":"قدمت AlloyDB من Google Cloud في النسخة التجريبية محرك HNSW المعزز بالمحرك العمودي، مما جعل عدد استعلامات البحث عن المتجهات باستخدام pgvector في الثانية (QPS) أعلى بما يصل إلى 4.9 مرات مقارنةً بـ HNSW في PostgreSQL القياسي، كما يمكن تحسين معدل الاسترجاع بمقدار 0.163 عند QPS ثابت. يتم تحقيق هذا التسريع عن طريق تثبيت الفهرس في ذاكرة المحرك العمودي، واستخدام تنسيق ذاكرة متجهة، وتجاوز تكلفة إدارة التخزين المؤقت القياسية في PostgreSQL، دون الحاجة إلى تعديل التطبيق للحصول على تحسن في الأداء.","category":"产品更新","source":"Google Cloud：Databases（RSS）","aggregationSource":"Google Cloud：Databases（RSS）","pageTitle":"أطلقت AlloyDB محرك أعمدة لتسريع HNSW، وزادت سرعة بحث المتجهات pgvector QPS حتى 4.9 مرات - Aioga أخبار الذكاء الاصطناعي","description":"قدمت AlloyDB من Google Cloud في النسخة التجريبية محرك HNSW المعزز بالمحرك العمودي، مما جعل عدد استعلامات البحث عن المتجهات باستخدام pgvector في الثانية (QPS) أعلى بما يصل إلى 4.9 م...","url":"https://www.aioga.com/ar/news/cms3dqo6s0axdro3fg3b4mfc0/","contentTranslated":true,"sourceHash":"30fec604009e3a2d","translatedAt":"2026-07-27T15:49:42.748Z"},"hi":{"title":"AlloyDB ने कॉलम-स्टोर इंजन के साथ HNSW को तेज किया, pgvector वेक्टर खोज QPS अधिकतम 4.9 गुना बढ़ी","summary":"Google Cloud का AlloyDB प्रीव्यू संस्करण में कॉलम-आधारित इंजन-संचालित HNSW इंडेक्स पेश करता है, जिससे pgvector वेक्टर सर्च के प्रति सेकंड क्वेरी (QPS) में मानक PostgreSQL HNSW की तुलना में अधिकतम 4.9 गुना की वृद्धि होती है, और स्थिर QPS पर रिकॉल रेट 0.163 तक बढ़ सकता है। यह संवर्द्धन इंडेक्स को कॉलम-आधारित इंजन की मेमोरी में फ़िक्स करके, वेक्टराइज्ड मेमोरी लेआउट का उपयोग करके और मानक PostgreSQL बफ़र प्रबंधक के ओवरहेड को बायपास करके प्राप्त किया जाता है, बिना किसी एप्लिकेशन संशोधन के प्रदर्शन में सुधार करता है।","category":"产品更新","source":"Google Cloud：Databases（RSS）","aggregationSource":"Google Cloud：Databases（RSS）","pageTitle":"AlloyDB ने कॉलम-स्टोर इंजन के साथ HNSW को तेज किया, pgvector वेक्टर खोज QPS अधिकतम 4.9 गुना बढ़ी - Aioga AI समाचार","description":"Google Cloud का AlloyDB प्रीव्यू संस्करण में कॉलम-आधारित इंजन-संचालित HNSW इंडेक्स पेश करता है, जिससे pgvector वेक्टर सर्च के प्रति सेकंड क्वेरी (QPS) में मानक PostgreSQL HNSW की त...","url":"https://www.aioga.com/hi/news/cms3dqo6s0axdro3fg3b4mfc0/","contentTranslated":true,"sourceHash":"30fec604009e3a2d","translatedAt":"2026-07-27T15:49:42.116Z"},"it":{"title":"AlloyDB lancia il motore a colonne per accelerare HNSW, la ricerca di vettori pgvector aumenta fino a 4,9 volte il QPS","summary":"AlloyDB di Google Cloud ha lanciato nella versione anteprima un indice HNSW accelerato da motore a colonne, che consente alla ricerca vettoriale pgvector di aumentare fino a 4,9 volte il numero di query al secondo (QPS) rispetto al normale HNSW di PostgreSQL, e di migliorare il tasso di richiamo di 0,163 mantenendo un QPS fisso. Questa accelerazione si ottiene fissando l'indice nella memoria del motore a colonne, utilizzando un layout di memoria vettorializzato e bypassando l'overhead del gestore di buffer standard di PostgreSQL, senza bisogno di modificare l'applicazione per ottenere miglioramenti delle prestazioni.","category":"产品更新","source":"Google Cloud：Databases（RSS）","aggregationSource":"Google Cloud：Databases（RSS）","pageTitle":"AlloyDB lancia il motore a colonne per accelerare HNSW, la ricerca di vettori pgvector aumenta fino a 4,9 volte il QPS - Aioga Notizie IA","description":"AlloyDB di Google Cloud ha lanciato nella versione anteprima un indice HNSW accelerato da motore a colonne, che consente alla ricerca vettoriale pgvector di aumentare fino a 4,9 vo...","url":"https://www.aioga.com/it/news/cms3dqo6s0axdro3fg3b4mfc0/","contentTranslated":true,"sourceHash":"30fec604009e3a2d","translatedAt":"2026-07-27T15:50:34.584Z"},"nl":{"title":"AlloyDB introduceert kolomgebaseerde engine voor het versnellen van HNSW, pgvector vectorgestuurde zoek-QPS tot 4,9 keer verhoogd","summary":"Google Cloud's AlloyDB heeft in de preview-versie een kolomgebaseerde engine-versnelde HNSW-index geïntroduceerd, waardoor het aantal queries per seconde (QPS) bij pgvector-vectorzoekopdrachten tot 4,9 keer kan toenemen vergeleken met standaard PostgreSQL HNSW, terwijl de terugroeping bij een vaste QPS met 0,163 kan verbeteren. Deze versnelling wordt bereikt door de index in het geheugen van de kolomgebaseerde engine te verankeren, gebruik te maken van een gevectoriseerde geheugenindeling en de overhead van de standaard PostgreSQL-bufferbeheerder te omzeilen, waardoor prestatieverbeteringen kunnen worden behaald zonder de applicatie te hoeven wijzigen.","category":"产品更新","source":"Google Cloud：Databases（RSS）","aggregationSource":"Google Cloud：Databases（RSS）","pageTitle":"AlloyDB introduceert kolomgebaseerde engine voor het versnellen van HNSW, pgvector vectorgestuurde zoek-QPS tot 4,9 keer verhoogd - Aioga AI-nieuws","description":"Google Cloud's AlloyDB heeft in de preview-versie een kolomgebaseerde engine-versnelde HNSW-index geïntroduceerd, waardoor het aantal queries per seconde (QPS) bij pgvector-vectorz...","url":"https://www.aioga.com/nl/news/cms3dqo6s0axdro3fg3b4mfc0/","contentTranslated":true,"sourceHash":"30fec604009e3a2d","translatedAt":"2026-07-27T15:50:38.579Z"},"tr":{"title":"AlloyDB, sütunlu motor ile HNSW hızlandırmasını tanıttı, pgvector vektör arama QPS'yi en fazla 4,9 kat artırdı","summary":"Google Cloud'un AlloyDB'si, önizleme sürümünde sütun bazlı motorla hızlandırılmış HNSW indeksini sundu; bu sayede pgvector vektör aramasının saniye başına sorgu sayısı (QPS), standart PostgreSQL HNSW'ye kıyasla en fazla 4,9 kat artarken, sabit QPS altında geri çağırma oranı 0,163 yükseldi. Bu hızlandırma, indeksin sütun bazlı motor belleğinde sabitlenmesi, vektörleştirilmiş bellek düzeninin kullanılması ve standart PostgreSQL tampon yöneticisi maliyetinin atlanmasıyla sağlanmakta olup, uygulamayı değiştirmeden performans artışı elde ediliyor.","category":"产品更新","source":"Google Cloud：Databases（RSS）","aggregationSource":"Google Cloud：Databases（RSS）","pageTitle":"AlloyDB, sütunlu motor ile HNSW hızlandırmasını tanıttı, pgvector vektör arama QPS'yi en fazla 4,9 kat artırdı - Aioga AI Haberleri","description":"Google Cloud'un AlloyDB'si, önizleme sürümünde sütun bazlı motorla hızlandırılmış HNSW indeksini sundu; bu sayede pgvector vektör aramasının saniye başına sorgu sayısı (QPS), stand...","url":"https://www.aioga.com/tr/news/cms3dqo6s0axdro3fg3b4mfc0/","contentTranslated":true,"sourceHash":"30fec604009e3a2d","translatedAt":"2026-07-27T15:51:21.373Z"},"vi":{"title":"AlloyDB ra mắt động cơ cột tăng tốc HNSW, tìm kiếm vector pgvector QPS tăng tối đa 4,9 lần","summary":"AlloyDB của Google Cloud đã ra mắt chỉ mục HNSW tăng tốc bằng bộ nhớ cột trong bản xem trước, giúp cho số truy vấn mỗi giây (QPS) của tìm kiếm vector pgvector so với HNSW tiêu chuẩn của PostgreSQL tăng lên đến 4,9 lần, và tỷ lệ gọi lại ở QPS cố định có thể tăng 0,163. Tăng tốc này được thực hiện bằng cách cố định chỉ mục trong bộ nhớ của bộ nhớ cột, sử dụng bố cục bộ nhớ vector hóa và bỏ qua chi phí quản lý bộ đệm tiêu chuẩn của PostgreSQL, không cần sửa đổi ứng dụng cũng có thể cải thiện hiệu suất.","category":"产品更新","source":"Google Cloud：Databases（RSS）","aggregationSource":"Google Cloud：Databases（RSS）","pageTitle":"AlloyDB ra mắt động cơ cột tăng tốc HNSW, tìm kiếm vector pgvector QPS tăng tối đa 4,9 lần - Tin tức AI Aioga","description":"AlloyDB của Google Cloud đã ra mắt chỉ mục HNSW tăng tốc bằng bộ nhớ cột trong bản xem trước, giúp cho số truy vấn mỗi giây (QPS) của tìm kiếm vector pgvector so với HNSW tiêu chuẩ...","url":"https://www.aioga.com/vi/news/cms3dqo6s0axdro3fg3b4mfc0/","contentTranslated":true,"sourceHash":"30fec604009e3a2d","translatedAt":"2026-07-27T15:51:21.122Z"},"id":{"title":"AlloyDB meluncurkan mesin kolom untuk mempercepat HNSW, pencarian vektor pgvector meningkatkan QPS hingga 4,9 kali","summary":"AlloyDB Google Cloud meluncurkan indeks HNSW yang dipercepat mesin kolom dalam versi pratinjau, sehingga jumlah kueri per detik (QPS) pencarian vektor pgvector meningkat hingga 4,9 kali dibandingkan HNSW PostgreSQL standar, dan rasio recall bisa meningkat 0,163 pada QPS yang tetap. Percepatan ini dicapai dengan menempatkan indeks di memori mesin kolom, menggunakan tata letak memori vektorisasi, dan melewati overhead manajer buffer PostgreSQL standar, sehingga peningkatan kinerja dapat diperoleh tanpa mengubah aplikasi.","category":"产品更新","source":"Google Cloud：Databases（RSS）","aggregationSource":"Google Cloud：Databases（RSS）","pageTitle":"AlloyDB meluncurkan mesin kolom untuk mempercepat HNSW, pencarian vektor pgvector meningkatkan QPS hingga 4,9 kali - Berita AI Aioga","description":"AlloyDB Google Cloud meluncurkan indeks HNSW yang dipercepat mesin kolom dalam versi pratinjau, sehingga jumlah kueri per detik (QPS) pencarian vektor pgvector meningkat hingga 4,9...","url":"https://www.aioga.com/id/news/cms3dqo6s0axdro3fg3b4mfc0/","contentTranslated":true,"sourceHash":"30fec604009e3a2d","translatedAt":"2026-07-27T15:52:10.189Z"},"th":{"title":"AlloyDB เปิดตัวเอนจินแบบคอลัมน์เพื่อเร่ง HNSW การค้นหาด้วยเวกเตอร์ pgvector เพิ่ม QPS สูงสุด 4.9 เท่า","summary":"Google Cloud AlloyDB เปิดตัวดัชนี HNSW ที่เร่งความเร็วด้วยเอนจินแบบคอลัมน์ในรุ่นพรีวิว ทำให้การค้นหาเวกเตอร์ pgvector ต่อวินาที (QPS) เพิ่มขึ้นสูงสุด 4.9 เท่าเมื่อเทียบกับ HNSW ของ PostgreSQL มาตรฐาน และเมื่อ QPS คงที่ อัตราการเรียกคืนสามารถเพิ่มขึ้น 0.163 การเร่งความเร็วนี้ทำได้โดยการล็อกดัชนีไว้ในหน่วยความจำของเอนจินแบบคอลัมน์ ใช้รูปแบบหน่วยความจำแบบเวกเตอร์ และหลีกเลี่ยงค่าใช้จ่ายของตัวจัดการบัฟเฟอร์มาตรฐานของ PostgreSQL ทำให้ไม่ต้องแก้ไขแอปพลิเคชันก็สามารถได้รับประสิทธิภาพที่สูงขึ้น","category":"产品更新","source":"Google Cloud：Databases（RSS）","aggregationSource":"Google Cloud：Databases（RSS）","pageTitle":"AlloyDB เปิดตัวเอนจินแบบคอลัมน์เพื่อเร่ง HNSW การค้นหาด้วยเวกเตอร์ pgvector เพิ่ม QPS สูงสุด 4.9 เท่า - ข่าว AI Aioga","description":"Google Cloud AlloyDB เปิดตัวดัชนี HNSW ที่เร่งความเร็วด้วยเอนจินแบบคอลัมน์ในรุ่นพรีวิว ทำให้การค้นหาเวกเตอร์ pgvector ต่อวินาที (QPS) เพิ่มขึ้นสูงสุด 4.9 เท่าเมื่อเทียบกับ HNSW ของ...","url":"https://www.aioga.com/th/news/cms3dqo6s0axdro3fg3b4mfc0/","contentTranslated":true,"sourceHash":"30fec604009e3a2d","translatedAt":"2026-07-27T15:52:12.377Z"},"pl":{"title":"AlloyDB wprowadza kolumnowy silnik przyspieszający HNSW, wyszukiwanie wektorów pgvector zwiększa QPS nawet 4,9 razy","summary":"AlloyDB w Google Cloud w wersji podglądowej wprowadził indeks HNSW przyspieszony przez silnik kolumnowy, dzięki czemu liczba zapytań na sekundę (QPS) w wyszukiwaniu wektorów pgvector wzrasta nawet 4,9 razy w porównaniu ze standardowym HNSW w PostgreSQL, a przy stałym QPS współczynnik odzyskania może zwiększyć się o 0,163. To przyspieszenie osiągane jest poprzez utrzymywanie indeksu w pamięci silnika kolumnowego, stosowanie wektoryzowanego układu pamięci oraz omijanie kosztów standardowego menedżera buforów PostgreSQL, co pozwala uzyskać wzrost wydajności bez konieczności modyfikowania aplikacji.","category":"产品更新","source":"Google Cloud：Databases（RSS）","aggregationSource":"Google Cloud：Databases（RSS）","pageTitle":"AlloyDB wprowadza kolumnowy silnik przyspieszający HNSW, wyszukiwanie wektorów pgvector zwiększa QPS nawet 4,9 razy - Aioga Wiadomości AI","description":"AlloyDB w Google Cloud w wersji podglądowej wprowadził indeks HNSW przyspieszony przez silnik kolumnowy, dzięki czemu liczba zapytań na sekundę (QPS) w wyszukiwaniu wektorów pgvect...","url":"https://www.aioga.com/pl/news/cms3dqo6s0axdro3fg3b4mfc0/","contentTranslated":true,"sourceHash":"30fec604009e3a2d","translatedAt":"2026-07-27T15:53:01.475Z"}}}}