Editing article

Title

Summary

Content <strong class="release-note-product-title">AlloyDB for PostgreSQL</strong> <h3>Feature</h3> <p>AlloyDB Language Connectors are now generally available (<a href="https://cloud.google.com/products#product-launch-stages">GA</a>). These language connectors are libraries that provide automated mutual TLS connections, IAM-based authorization, and Automated IAM Authentication when connecting to an AlloyDB instance. For more information about language connectors, see <a href="https://cloud.google.com/alloydb/docs/language-connectors-overview">AlloyDB Language Connectors overview</a>.</p> <strong class="release-note-product-title">Anti Money Laundering AI</strong> <h3>Announcement</h3> <p>Added a new metric to AML AI, providing insight into the importance of each feature family to an AML AI Model. This metric is available in new v4 <a href="https://cloud.google.com/financial-services/anti-money-laundering/docs/reference/engine-versions">engine versions</a>. It allows you to:</p> <ul> <li>Act on model monitoring outputs in the context of their importance to a model</li> <li>Check the contribution of your Party Supplementary Data to a model</li> </ul> <strong class="release-note-product-title">App Engine standard environment Go</strong> <h3>Breaking</h3> <p>You can&#39;t use the latest version of <code>dev_appserver.py</code> to locally run your applications for runtimes that reached <a href="https://cloud.google.com/appengine/docs/standard/lifecycle/support-schedule">end of support</a>. To continue using an archived version of <code>dev_appserver.py</code>, see <a href="https://cloud.google.com/appengine/docs/standard/tools/using-local-server-for-eos-runtimes">Use the local development server after runtimes reach the end of support</a>.</p> <strong class="release-note-product-title">App Engine standard environment Java</strong> <h3>Breaking</h3> <p>You can&#39;t use the latest version of <code>dev_appserver.py</code> to locally run your applications for runtimes that reached <a href="https://cloud.google.com/appengine/docs/standard/lifecycle/support-schedule">end of support</a>. To continue using an archived version of <code>dev_appserver.py</code>, see <a href="https://cloud.google.com/appengine/docs/standard/tools/using-local-server-for-eos-runtimes">Use the local development server after runtimes reach the end of support</a>.</p> <strong class="release-note-product-title">App Engine standard environment PHP</strong> <h3>Breaking</h3> <p>You can&#39;t use the latest version of <code>dev_appserver.py</code> to locally run your applications for runtimes that reached <a href="https://cloud.google.com/appengine/docs/standard/lifecycle/support-schedule">end of support</a>. To continue using an archived version of <code>dev_appserver.py</code>, see <a href="https://cloud.google.com/appengine/docs/standard/tools/using-local-server-for-eos-runtimes">Use the local development server after runtimes reach the end of support</a>.</p> <strong class="release-note-product-title">App Engine standard environment Python</strong> <h3>Breaking</h3> <p>You can&#39;t use the latest version of <code>dev_appserver.py</code> to locally run your applications for runtimes that reached <a href="https://cloud.google.com/appengine/docs/standard/lifecycle/support-schedule">end of support</a>. To continue using an archived version of <code>dev_appserver.py</code>, see <a href="https://cloud.google.com/appengine/docs/standard/tools/using-local-server-for-eos-runtimes">Use the local development server after runtimes reach the end of support</a>.</p> <strong class="release-note-product-title">Blockchain Node Engine</strong> <h3>Announcement</h3> <p>On March 12, 2024, Blockchain Node Engine upgraded all mainnet Ethereum nodes in preparation for the Dencun Hardfork.</p> <strong class="release-note-product-title">Certificate Manager</strong> <h3>Feature</h3> <p>Certificate Manager supports integration with regional external Application Load Balancers and regional internal Application Load Balancers. This feature is generally available (<a href="https://cloud.google.com/products#product-launch-stages">GA</a>). For more information, see <a href="https://cloud.google.com/certificate-manager/docs/overview">Certificate Manager overview</a>.</p> <strong class="release-note-product-title">Chronicle</strong> <h3>Feature</h3> <p><a href="https://cloud.google.com/chronicle/docs/install/troubleshoot-forwarder">Forwarder troubleshooting guide</a> is now available to help you diagnose and resolve common issues that may arise while using the Chronicle Linux forwarder. </p> <strong class="release-note-product-title">Cloud Composer</strong> <h3>Fixed</h3> <p>Fixed creation and upgrades in environments that have environment variables with special symbols.</p> <h3>Changed</h3> <p><strong>Cloud Composer 2.6.4</strong> <a href="https://cloud.google.com/composer/docs/concepts/versioning/composer-versions">images</a> are available:</p> <ul> <li>composer-2.6.4-airflow-2.6.3 (default)</li> <li>composer-2.6.4-airflow-2.5.3</li> </ul> <h3>Deprecated</h3> <p>Cloud Composer versions 2.1.8 and 1.20.8 have reached their <a href="https://cloud.google.com/composer/docs/concepts/versioning/composer-versioning-overview#version-deprecation-and-support">end of full support period</a>.</p> <strong class="release-note-product-title">Cloud Data Fusion</strong> <h3>Fixed</h3> <p>The Cloud Data Fusion version 6.9.2.3 patch revision is generally available (<a href="https://cloud.google.com/products#product-launch-stages">GA</a>). 6.9.2.3 includes the following fixes:</p> <ul> <li><p>Skipped running <code>MetadataConsumerSubscriberService</code> when <a href="https://cloud.google.com/data-fusion/docs/how-to/view-lineage-in-dataplex">Dataplex Data Lineage Integration</a> is disabled (<a href="https://cdap.atlassian.net/browse/CDAP-20947">CDAP-20947</a>).</p></li> <li><p>Fixed an issue causing runtime arguments of pipeline triggers to not propagate to downstream pipelines (<a href="https://cdap.atlassian.net/browse/CDAP-20943">CDAP-20943</a>).</p></li> <li><p>Fixed an issue causing pipelines to fail in starting state when the system worker service is intermittently unavailable (<a href="https://cdap.atlassian.net/browse/CDAP-20956">CDAP-20956</a>).</p></li> <li><p>Fixed an issue causing pipelines to fail in starting state when the Compute Engine metadata server is intermittently unavailable (<a href="https://cdap.atlassian.net/jira/software/c/projects/CDAP/issues/CDAP-20955">CDAP-20955</a>).</p></li> </ul> <strong class="release-note-product-title">Cloud Load Balancing</strong> <h3>Feature</h3> <p>The global external Proxy Network Load Balancer is implemented on globally distributed GFEs and supports advanced traffic management capabilities. This load balancer can be configured to handle either TCP or SSL traffic by using either a target TCP proxy or a target SSL proxy respectively. Global external proxy Network Load Balancers support backends such as instance groups, hybrid NEGs, and Private Service Connect NEGs. For details, see the <a href="https://cloud.google.com/load-balancing/docs/tcp">External proxy Network Load Balancer overview</a>.</p> <p>To set up a global external Proxy Network Load Balancer, see the following pages:</p> <ul> <li><a href="https://cloud.google.com/load-balancing/docs/tcp/set-up-global-ext-proxy-ssl">Instance group backends with SSL</a></li> <li><a href="https://cloud.google.com/load-balancing/docs/tcp/set-up-global-ext-proxy-tcp">Instance group backends with TCP</a></li> </ul> <p>This capability is in <strong>General Availability</strong>.</p> <h3>Feature</h3> <p>Regional external Application Load Balancers and regional internal Application Load Balancers now support Certificate Manager certificates. For more information, see <a href="https://cloud.google.com/load-balancing/docs/ssl-certificates#certificate-summary">Certificates and Google Cloud load balancers</a>.</p> <p>This capability is in <strong>General Availability</strong>.</p> <strong class="release-note-product-title">Cloud Storage</strong> <h3>Feature</h3> <p>You can now view granular bucket-level cost data in the <a href="https://cloud.google.com/billing/docs/how-to/export-data-bigquery-tables/detailed-usage">Cloud Billing Detailed data export</a>.</p> <strong class="release-note-product-title">Cloud Translation</strong> <h3>Feature</h3> <p>For AutoML datasets, you can <a href="https://cloud.google.com/translate/docs/advanced/automl-datasets#import-data">tag segment pairs</a> when importing them through the Google Cloud console.</p> <h3>Changed</h3> <p>You can request document translations with <a href="https://cloud.google.com/translate/docs/advanced/endpoints">multi-regional endpoints</a>.</p> <strong class="release-note-product-title">Compute Engine</strong> <h3>Feature</h3> <p><strong>Generally available</strong>: You can scale a single VM into a managed instance group (MIG), which is a group of VMs that you can manage as a single entity. A MIG can make your workload scalable and highly available using features like autoscaling, autohealing, regional (multiple zones) deployment, and automatic updating. </p> <p>For more information, see <a href="https://cloud.google.com/compute/docs/instance-groups/create-mig-from-vm">Create a MIG from an existing VM</a>.</p> <strong class="release-note-product-title">Container Registry</strong> <h3>Announcement</h3> <p>New tooling is available to upgrade from Container Registry to Artifact Registry. For more information on the available tools, see <a href="https://cloud.google.com/artifact-registry/docs/transition/auto-migrate-gcr-ar">Automate migration to Artifact Registry</a>.</p> <strong class="release-note-product-title">Dataform</strong> <h3>Feature</h3> <p>Granting repository access to all authenticated users is available. For more information, see <a href="https://cloud.google.com/dataform/docs/access-control#grant_public_access_to_a_repository">Grant public access to a repository</a>.</p> <strong class="release-note-product-title">Retail API</strong> <h3>Changed</h3> <p><strong>Vertex AI Search for retail: Renamed in the console and documentation</strong></p> <p>The Google Cloud console has been updated to show the current product name for Vertex AI Search for retail.</p> <p>You might see the old names (Retail or Retail API) in some places—for example, in the documentation. Google is in the process of updating content to reflect the new branding.</p> <strong class="release-note-product-title">Vertex AI Search and Conversation</strong> <h3>Announcement</h3> <p><strong>Vertex AI Search: Vertex AI Search for healthcare (GA)</strong></p> <p>Vertex AI Search for healthcare is Generally available (GA). Healthcare search lets you query healthcare records stored in FHIR data stores. For more information, see <a href="https://cloud.google.com/generative-ai-app-builder/docs/introduction#es-and-personalize">Vertex AI Search</a>. With healthcare search, you can:</p> <ul> <li><a href="https://cloud.google.com/generative-ai-app-builder/docs/create-data-store-hc">Create a healthcare search data store and import FHIR R4 data</a></li> <li><a href="https://cloud.google.com/generative-ai-app-builder/docs/create-app-hc">Create a healthcare search app</a></li> <li><a href="https://cloud.google.com/generative-ai-app-builder/docs/search-hc-data">Search healthcare data using keywords and natural language queries</a></li> <li><a href="https://cloud.google.com/generative-ai-app-builder/docs/search-hc-data#nl-search-genAI">Search healthcare data using natural language queries with generative AI answers</a> (Private preview)</li> </ul> <h3>Feature</h3> <p><strong>Vertex AI Search: Specify a parser for unstructured content (Public preview)</strong></p> <p>You can control how documents are parsed when they are uploaded to Vertex AI Search. Parser specification is available in Public preview.</p> <p>Vertex AI Search provides a digital parser (GA), an OCR parser for PDFs (Public preview), and a layout parser (Public Preview). During data store creation for generic search apps with unstructured data, you can set a default parser for the data store and an override parser for specific file types.</p> <p>For more information, see <a href="https://cloud.google.com/generative-ai-app-builder/docs/parse-chunk-documents#parsing">Parse documents</a>. </p> <h3>Feature</h3> <p><strong>Vertex AI Search: Turn on document chunking (Public preview)</strong></p> <p>To use Vertex AI Search for retrieval-augmented generation (RAG) for LLMs, you can turn on document chunking when creating a data store. Document chunking is available in Public preview.</p> <p>When document chunking is turned on, your documents are split into chunks when you ingest documents into your data store, and your search app can return chunks of data in search results instead of full documents. Using chunked data for RAG increases relevance for LLM answers and reduces computational load for LLMs. Document chunking is in Public preview. For more information, see <a href="https://cloud.google.com/generative-ai-app-builder/docs/parse-chunk-documents#parse-chunk-rag">Chunk documents for RAG</a>. </p> <h3>Feature</h3> <p><strong>Vertex AI Search: Connect ServiceNow as a data source (Private preview)</strong></p> <p>You can connect ServiceNow as a third-party data source for Vertex AI Search. For more information, see <a href="https://cloud.google.com/generative-ai-app-builder/docs/connect-third-party-data-source">Connect a third-party data source</a>.</p>

Author

Link

Published date

Image url

Feed url

Guid

Hidden blurb --- !ruby/object:Feedjira::Parser::AtomEntry entry_id: tag:google.com,2016:gcp-release-notes#March_12_2024 content: "<strong class=\"release-note-product-title\">AlloyDB for PostgreSQL</strong>\n<h3>Feature</h3>\n<p>AlloyDB Language Connectors are now generally available (<a href=\"https://cloud.google.com/products#product-launch-stages\">GA</a>). These language connectors are libraries that provide automated mutual TLS connections, IAM-based authorization, and Automated IAM Authentication when connecting to an AlloyDB instance. For more information about language connectors, see <a href=\"https://cloud.google.com/alloydb/docs/language-connectors-overview\">AlloyDB Language Connectors overview</a>.</p>\n<strong class=\"release-note-product-title\">Anti Money Laundering AI</strong>\n<h3>Announcement</h3>\n<p>Added a new metric to AML AI, providing insight into the importance of each feature family to an AML AI Model. This metric is available in new v4 <a href=\"https://cloud.google.com/financial-services/anti-money-laundering/docs/reference/engine-versions\">engine versions</a>. It allows you to:</p>\n\n<ul>\n<li>Act on model monitoring outputs in the context of their importance to a model</li>\n<li>Check the contribution of your Party Supplementary Data to a model</li>\n</ul>\n<strong class=\"release-note-product-title\">App Engine standard environment Go</strong>\n<h3>Breaking</h3>\n<p>You can&#39;t use the latest version of <code>dev_appserver.py</code> to locally run your\napplications for runtimes that reached <a href=\"https://cloud.google.com/appengine/docs/standard/lifecycle/support-schedule\">end of support</a>.\nTo continue using an archived version of <code>dev_appserver.py</code>, see\n<a href=\"https://cloud.google.com/appengine/docs/standard/tools/using-local-server-for-eos-runtimes\">Use the local development server after runtimes reach the end of support</a>.</p>\n<strong class=\"release-note-product-title\">App Engine standard environment Java</strong>\n<h3>Breaking</h3>\n<p>You can&#39;t use the latest version of <code>dev_appserver.py</code> to locally run your\napplications for runtimes that reached <a href=\"https://cloud.google.com/appengine/docs/standard/lifecycle/support-schedule\">end of support</a>.\nTo continue using an archived version of <code>dev_appserver.py</code>, see\n<a href=\"https://cloud.google.com/appengine/docs/standard/tools/using-local-server-for-eos-runtimes\">Use the local development server after runtimes reach the end of support</a>.</p>\n<strong class=\"release-note-product-title\">App Engine standard environment PHP</strong>\n<h3>Breaking</h3>\n<p>You can&#39;t use the latest version of <code>dev_appserver.py</code> to locally run your\napplications for runtimes that reached <a href=\"https://cloud.google.com/appengine/docs/standard/lifecycle/support-schedule\">end of support</a>.\nTo continue using an archived version of <code>dev_appserver.py</code>, see\n<a href=\"https://cloud.google.com/appengine/docs/standard/tools/using-local-server-for-eos-runtimes\">Use the local development server after runtimes reach the end of support</a>.</p>\n<strong class=\"release-note-product-title\">App Engine standard environment Python</strong>\n<h3>Breaking</h3>\n<p>You can&#39;t use the latest version of <code>dev_appserver.py</code> to locally run your\napplications for runtimes that reached <a href=\"https://cloud.google.com/appengine/docs/standard/lifecycle/support-schedule\">end of support</a>.\nTo continue using an archived version of <code>dev_appserver.py</code>, see\n<a href=\"https://cloud.google.com/appengine/docs/standard/tools/using-local-server-for-eos-runtimes\">Use the local development server after runtimes reach the end of support</a>.</p>\n<strong class=\"release-note-product-title\">Blockchain Node Engine</strong>\n<h3>Announcement</h3>\n<p>On March 12, 2024, Blockchain Node Engine upgraded all mainnet Ethereum nodes in preparation for the Dencun Hardfork.</p>\n<strong class=\"release-note-product-title\">Certificate Manager</strong>\n<h3>Feature</h3>\n<p>Certificate Manager supports integration with regional external Application Load Balancers and regional internal Application Load Balancers. This feature is generally available (<a href=\"https://cloud.google.com/products#product-launch-stages\">GA</a>). For more information, see <a href=\"https://cloud.google.com/certificate-manager/docs/overview\">Certificate Manager overview</a>.</p>\n<strong class=\"release-note-product-title\">Chronicle</strong>\n<h3>Feature</h3>\n<p><a href=\"https://cloud.google.com/chronicle/docs/install/troubleshoot-forwarder\">Forwarder troubleshooting guide</a> is now available to help you diagnose and resolve common issues that may arise while using the Chronicle Linux forwarder. </p>\n<strong class=\"release-note-product-title\">Cloud Composer</strong>\n<h3>Fixed</h3>\n<p>Fixed creation and upgrades in environments that have environment variables with special symbols.</p>\n<h3>Changed</h3>\n<p><strong>Cloud Composer 2.6.4</strong> <a href=\"https://cloud.google.com/composer/docs/concepts/versioning/composer-versions\">images</a> are available:</p>\n\n<ul>\n<li>composer-2.6.4-airflow-2.6.3 (default)</li>\n<li>composer-2.6.4-airflow-2.5.3</li>\n</ul>\n<h3>Deprecated</h3>\n<p>Cloud Composer versions 2.1.8 and 1.20.8 have reached their <a href=\"https://cloud.google.com/composer/docs/concepts/versioning/composer-versioning-overview#version-deprecation-and-support\">end of full support period</a>.</p>\n<strong class=\"release-note-product-title\">Cloud Data Fusion</strong>\n<h3>Fixed</h3>\n<p>The Cloud Data Fusion version 6.9.2.3 patch revision is generally available (<a href=\"https://cloud.google.com/products#product-launch-stages\">GA</a>). 6.9.2.3 includes the following fixes:</p>\n\n<ul>\n<li><p>Skipped running <code>MetadataConsumerSubscriberService</code> when <a href=\"https://cloud.google.com/data-fusion/docs/how-to/view-lineage-in-dataplex\">Dataplex Data Lineage Integration</a> is disabled (<a href=\"https://cdap.atlassian.net/browse/CDAP-20947\">CDAP-20947</a>).</p></li>\n<li><p>Fixed an issue causing runtime arguments of pipeline triggers to not propagate to downstream pipelines (<a href=\"https://cdap.atlassian.net/browse/CDAP-20943\">CDAP-20943</a>).</p></li>\n<li><p>Fixed an issue causing pipelines to fail in starting state when the system worker service is intermittently unavailable (<a href=\"https://cdap.atlassian.net/browse/CDAP-20956\">CDAP-20956</a>).</p></li>\n<li><p>Fixed an issue causing pipelines to fail in starting state when the Compute Engine metadata server is intermittently unavailable (<a href=\"https://cdap.atlassian.net/jira/software/c/projects/CDAP/issues/CDAP-20955\">CDAP-20955</a>).</p></li>\n</ul>\n<strong class=\"release-note-product-title\">Cloud Load Balancing</strong>\n<h3>Feature</h3>\n<p>The global external Proxy Network Load Balancer is implemented on globally distributed GFEs and supports advanced traffic management capabilities. This load balancer can be configured to handle either TCP or SSL traffic by using either a target TCP proxy or a target SSL proxy respectively. Global external proxy Network Load Balancers support backends such as instance groups, hybrid NEGs, and Private Service Connect NEGs.\nFor details, see the <a href=\"https://cloud.google.com/load-balancing/docs/tcp\">External proxy Network Load Balancer overview</a>.</p>\n\n<p>To set up a global external Proxy Network Load Balancer, see the following pages:</p>\n\n<ul>\n<li><a href=\"https://cloud.google.com/load-balancing/docs/tcp/set-up-global-ext-proxy-ssl\">Instance group backends with SSL</a></li>\n<li><a href=\"https://cloud.google.com/load-balancing/docs/tcp/set-up-global-ext-proxy-tcp\">Instance group backends with TCP</a></li>\n</ul>\n\n<p>This capability is in <strong>General Availability</strong>.</p>\n<h3>Feature</h3>\n<p>Regional external Application Load Balancers and regional internal Application Load Balancers now support Certificate Manager certificates. For more information, see <a href=\"https://cloud.google.com/load-balancing/docs/ssl-certificates#certificate-summary\">Certificates and Google Cloud load balancers</a>.</p>\n\n<p>This capability is in <strong>General Availability</strong>.</p>\n<strong class=\"release-note-product-title\">Cloud Storage</strong>\n<h3>Feature</h3>\n<p>You can now view granular bucket-level cost data in the <a href=\"https://cloud.google.com/billing/docs/how-to/export-data-bigquery-tables/detailed-usage\">Cloud Billing Detailed data export</a>.</p>\n<strong class=\"release-note-product-title\">Cloud Translation</strong>\n<h3>Feature</h3>\n<p>For AutoML datasets, you can <a href=\"https://cloud.google.com/translate/docs/advanced/automl-datasets#import-data\">tag segment pairs</a> when importing them through the Google Cloud console.</p>\n<h3>Changed</h3>\n<p>You can request document translations with <a href=\"https://cloud.google.com/translate/docs/advanced/endpoints\">multi-regional endpoints</a>.</p>\n<strong class=\"release-note-product-title\">Compute Engine</strong>\n<h3>Feature</h3>\n<p><strong>Generally available</strong>: You can scale a single VM into a managed instance group (MIG), which is a group of VMs that you can manage as a single entity. A MIG can make your workload scalable and highly available using features like autoscaling, autohealing, regional (multiple zones) deployment, and automatic updating. </p>\n\n<p>For more information, see <a href=\"https://cloud.google.com/compute/docs/instance-groups/create-mig-from-vm\">Create a MIG from an existing VM</a>.</p>\n<strong class=\"release-note-product-title\">Container Registry</strong>\n<h3>Announcement</h3>\n<p>New tooling is available to upgrade from Container Registry to Artifact Registry. For more information on the available tools, see <a href=\"https://cloud.google.com/artifact-registry/docs/transition/auto-migrate-gcr-ar\">Automate migration to Artifact Registry</a>.</p>\n<strong class=\"release-note-product-title\">Dataform</strong>\n<h3>Feature</h3>\n<p>Granting repository access to all authenticated users is available. For more information, see <a href=\"https://cloud.google.com/dataform/docs/access-control#grant_public_access_to_a_repository\">Grant public access to a repository</a>.</p>\n<strong class=\"release-note-product-title\">Retail API</strong>\n<h3>Changed</h3>\n<p><strong>Vertex AI Search for retail: Renamed in the console and documentation</strong></p>\n\n<p>The Google Cloud console has been updated to show the current product name for Vertex AI Search for retail.</p>\n\n<p>You might see the old names (Retail or Retail API) in some places—for example, in the documentation. Google is in the process of updating content to reflect the new branding.</p>\n<strong class=\"release-note-product-title\">Vertex AI Search and Conversation</strong>\n<h3>Announcement</h3>\n<p><strong>Vertex AI Search: Vertex AI Search for healthcare (GA)</strong></p>\n\n<p>Vertex AI Search for healthcare is Generally available (GA). Healthcare search lets you query healthcare records stored in FHIR data stores. For more information, see <a href=\"https://cloud.google.com/generative-ai-app-builder/docs/introduction#es-and-personalize\">Vertex AI Search</a>. With healthcare search, you can:</p>\n\n<ul>\n<li><a href=\"https://cloud.google.com/generative-ai-app-builder/docs/create-data-store-hc\">Create a healthcare search data store and import FHIR R4 data</a></li>\n<li><a href=\"https://cloud.google.com/generative-ai-app-builder/docs/create-app-hc\">Create a healthcare search app</a></li>\n<li><a href=\"https://cloud.google.com/generative-ai-app-builder/docs/search-hc-data\">Search healthcare data using keywords and natural language queries</a></li>\n<li><a href=\"https://cloud.google.com/generative-ai-app-builder/docs/search-hc-data#nl-search-genAI\">Search healthcare data using natural language queries with generative AI answers</a> (Private preview)</li>\n</ul>\n<h3>Feature</h3>\n<p><strong>Vertex AI Search: Specify a parser for unstructured content (Public preview)</strong></p>\n\n<p>You can control how documents are parsed when they are uploaded to Vertex AI Search. Parser specification is available in Public preview.</p>\n\n<p>Vertex AI Search provides a digital parser (GA), an OCR parser for PDFs (Public preview), and a layout parser (Public Preview). During data store creation for generic search apps with unstructured data, you can set a default parser for the data store and an override parser for specific file types.</p>\n\n<p>For more information, see <a href=\"https://cloud.google.com/generative-ai-app-builder/docs/parse-chunk-documents#parsing\">Parse documents</a>. </p>\n<h3>Feature</h3>\n<p><strong>Vertex AI Search: Turn on document chunking (Public preview)</strong></p>\n\n<p>To use Vertex AI Search for retrieval-augmented generation (RAG) for LLMs, you can turn on document chunking when creating a data store. Document chunking is available in Public preview.</p>\n\n<p>When document chunking is turned on, your documents are split into chunks when you ingest documents into your data store, and your search app can return chunks of data in search results instead of full documents. Using chunked data for RAG increases relevance for LLM answers and reduces computational load for LLMs. Document chunking is in Public preview. For more information, see <a href=\"https://cloud.google.com/generative-ai-app-builder/docs/parse-chunk-documents#parse-chunk-rag\">Chunk documents for RAG</a>. </p>\n<h3>Feature</h3>\n<p><strong>Vertex AI Search: Connect ServiceNow as a data source (Private preview)</strong></p>\n\n<p>You can connect ServiceNow as a third-party data source for Vertex AI Search. For more information, see <a href=\"https://cloud.google.com/generative-ai-app-builder/docs/connect-third-party-data-source\">Connect a third-party data source</a>.</p>\n\n " title_type: published: &1 2024-03-12 07:00:00.000000000 Z updated: *1 links: - https://cloud.google.com/release-notes#March_12_2024 title: March 12, 2024 carlessian_info: news_filer_version: 2 newspaper: GCP latest releases macro_region: Technology rss_fields: - entry_id - content - title_type - published - updated - links - title categories: [] url: https://cloud.google.com/release-notes#March_12_2024

Language

Active

Ricc internal notes Imported via /Users/ricc/git/gemini-news-crawler/webapp/db/seeds.d/import-feedjira.rb on 2024-03-31 23:42:30 +0200. Content is EMPTY here. Entried: entry_id,content,title_type,published,updated,links,title. TODO add Newspaper: filename = /Users/ricc/git/gemini-news-crawler/webapp/db/seeds.d/../../../crawler/out/feedjira/Technology/GCP latest releases/2024-03-12-March_12,_2024-v2.yaml

Ricc source

Show this article Back to articles