設定包括為 Vertex AI 特徵儲存庫 (舊版) 設定專案的相關資訊，以及使用 Vertex AI 特徵儲存庫 (舊版) 的必要權限。
設定專案
以下程序說明如何建立新專案並啟用 Vertex AI API。使用 Vertex AI 特徵儲存庫 (舊版) 時，必須啟用這項 API。如果您現有的專案已啟用 Vertex AI API，可以使用該專案，不必建立新專案。
Sign in to your Google Cloud account. If you're new to
Google Cloud,
create an account to evaluate how our products perform in
real-world scenarios. New customers also get $300 in free credits to
run, test, and deploy workloads.
In the Google Cloud console, on the project selector page,
select or create a Google Cloud project.
除了使用者權限外，Vertex AI 特徵儲存庫 (舊版) 也會代表您執行作業，例如存取來源資料。為此，Vertex AI 特徵儲存庫 (舊版) 會使用服務代理：
service-PROJECT_NUMBER@gcp-sa-aiplatform.iam.gserviceaccount.com。
根據預設，服務代理會授予 Vertex AI 特徵儲存庫 (舊版) 權限，存取特徵儲存庫所在專案中的來源資料。如果來源資料與 Feature Store 位於不同專案，您必須授予服務代理存取來源資料所在專案的權限。
Vertex AI 特徵儲存庫 (舊版) 會強制執行配額和限制，協助您設定用量限制來管理資源，並預防用量意外暴增的情況，進而保障 Google Cloud 使用者社群的權益。為避免遇到非預期的限制，請在「配額和限制」頁面中查看 Vertex AI 特徵儲存庫 (舊版) 配額。舉例來說，Vertex AI 特徵儲存庫 (舊版) 會針對線上服務節點數量和每分鐘可提出的線上服務要求數量設定配額。
[[["容易理解","easyToUnderstand","thumb-up"],["確實解決了我的問題","solvedMyProblem","thumb-up"],["其他","otherUp","thumb-up"]],[["難以理解","hardToUnderstand","thumb-down"],["資訊或程式碼範例有誤","incorrectInformationOrSampleCode","thumb-down"],["缺少我需要的資訊/範例","missingTheInformationSamplesINeed","thumb-down"],["翻譯問題","translationIssue","thumb-down"],["其他","otherDown","thumb-down"]],["上次更新時間：2025-09-04 (世界標準時間)。"],[],[],null,["# Setup includes information about setting up a project for\nVertex AI Feature Store (Legacy) and the required permissions for using\nVertex AI Feature Store (Legacy).\n\nConfigure project\n-----------------\n\nThe following procedure describes how to create a new project and enable the\nVertex AI API. This API is required to use\nVertex AI Feature Store (Legacy). If you already have an existing project with\nthe Vertex AI API enabled, you can use that project instead of\ncreating a new project.\n\n- Sign in to your Google Cloud account. If you're new to Google Cloud, [create an account](https://console.cloud.google.com/freetrial) to evaluate how our products perform in real-world scenarios. New customers also get $300 in free credits to run, test, and deploy workloads.\n- In the Google Cloud console, on the project selector page,\n select or create a Google Cloud project.\n\n [Go to project selector](https://console.cloud.google.com/projectselector2/home/dashboard)\n-\n [Verify that billing is enabled for your Google Cloud project](/billing/docs/how-to/verify-billing-enabled#confirm_billing_is_enabled_on_a_project).\n\n-\n\n\n Enable the Vertex AI API.\n\n\n [Enable the API](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com)\n\n- In the Google Cloud console, on the project selector page,\n select or create a Google Cloud project.\n\n [Go to project selector](https://console.cloud.google.com/projectselector2/home/dashboard)\n-\n [Verify that billing is enabled for your Google Cloud project](/billing/docs/how-to/verify-billing-enabled#confirm_billing_is_enabled_on_a_project).\n\n-\n\n\n Enable the Vertex AI API.\n\n\n [Enable the API](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com)\n\nVertex AI Feature Store (Legacy) service agent\n----------------------------------------------\n\nIn addition to user permissions, Vertex AI Feature Store (Legacy) acts on your\nbehalf to perform operations such as accessing source data. To do so,\nVertex AI Feature Store (Legacy) uses a service agent:\n`service-`\u003cvar class=\"readonly\" translate=\"no\"\u003ePROJECT_NUMBER\u003c/var\u003e`@gcp-sa-aiplatform.iam.gserviceaccount.com`.\nBy default, the service agent grants Vertex AI Feature Store (Legacy) access\nto source data in the same project where your featurestore is located. If the\nsource data is in a different project from your featurestore, you must grant the\nservice agent permission to access the project where the source data is\nlocated.\n\nFor more information, see [Grant Vertex AI service agents access to other\nresources](/vertex-ai/docs/general/access-control#grant_service_agents_access_to_other_resources).\n\nIAM permissions\n---------------\n\nVertex AI admins have Vertex AI Feature Store (Legacy) administrator\nprivileges. If you require more granularity, Vertex AI Feature Store (Legacy)\nprovides a set of predefined IAM roles. These roles provide\ndifferent sets of permissions that are based on the following personas:\n\nIT operations and DevOps\n: IT operations and DevOps manage Google Cloud resources and are responsible for\n creating featurestores and tuning their performance. You can use the\n `featurestoreAdmin` or `featurestoreInstanceCreator` role. The instance creator\n role lets you manage featurestores but prevents you from viewing data or\n writing data to the featurestores.\n\nData scientists and data engineers\n: Data scientists and data engineers create features and write data to\n featurestores. You can use the `featurestoreResourceEditor` role to\n manage entity types and features, and use the `featurestoreDataWriter` role to\n read and write feature values.\n\nML researchers and business analysts\n: ML researchers and business analysts search for features and export values for\n training models or making predictions; they don't need to create new features or\n write data. You can use the `featurestoreResourceViewer` role to search\n or browse for features and the `featurestoreDataViewer` role to read feature\n values.\n\nFor descriptions of each role and their associated permissions, see\n[Predefined roles for\nVertex AI](/vertex-ai/docs/general/access-control#predefined-roles).\n\nQuotas and limits\n-----------------\n\nVertex AI Feature Store (Legacy) enforces quotas and limits to help you manage\nresources by setting your own usage limits and to protect the community of\nGoogle Cloud users by preventing unforeseen spikes in usage. To prevent you from\nhitting unplanned constraints, review Vertex AI Feature Store (Legacy) quotas\non the [Quotas and limits](/vertex-ai/quotas#featurestore) page. For example,\nVertex AI Feature Store (Legacy) sets a quota on the number of online serving\nnodes and a quota on the number of online serving requests that you can make per\nminute.\n\nWhat's next\n-----------\n\n- Learn about [Manage featurestores](/vertex-ai/docs/featurestore/managing-featurestores).\n- Learn about [best practices](/vertex-ai/docs/featurestore/best-practices) for using Vertex AI Feature Store (Legacy)."]]
