在本文件中,雲端足跡最佳化策略著重於規劃及設計最佳化策略時,如何善用 Active Assist 產品組合。Google Cloud
定義願景並瞭解驅動因素
企業必須先定義要用來做為雲端足跡最佳化方法依據的問題,常見問題如下:
安全性
成效
成本最佳化
靈活性
企業目標
開始為 Active Assist 建議設計自動化管道時,請先定義企業目標,並為每個目標指派優先順序。接著,您可以將這些優先事項對應至藍圖,在 Google Cloud 貴機構中推出及擴大使用 Active Assist。
舉例來說,企業可能會想使用 Active Assist 建議,進行安全性與成本最佳化。不過,企業一開始可能會選擇投資,為 Active Assist 生成的安全相關最佳化建議建立自動化管道。之後,隨著企業使用 Active Assist 產品組合的經驗越來越豐富,自動化歷程也日趨成熟,可能會自動套用其他類型的最佳化建議,例如 VM 適當規模調整和閒置 VM 建議。
設計策略
企業必須明確定義程序,說明如何審查及執行 Active Assist 生成的建議。建議您採取分階段做法,以有條不紊的方式逐步提高自動化程度。企業在機構中採用 Active Assist 時,可以採取下列疊代式方法: Google Cloud
在第一階段,您可以使用「Recommendation Hub」(建議中心),在Google Cloud 控制台中查看 Active Assist 建議。您可以使用主控台方法來查看及實作建議。這種做法可協助您熟悉 Active Assist 建議,同時評估這些建議是否適用。這項工具也能協助您決定要優先處理哪些類別的最佳化建議。如下圖所示,最佳化建議中心會顯示各資源類別的建議,並提供各資源的相關詳細資料。
[[["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,["# Patterns for using Active Assist at scale\n=========================================\n\nThis document is the first part in a series that introduces architectural\npatterns that enterprises can use to optimize their cloud footprint at scale\nusing [Active Assist](/solutions/active-assist).\nThis document is intended for people in the following roles:\n\n- Enterprise architects\n- Engineering leads\n- People who work in security and create automation to optimize cloud security, performance, and manageability\n\nThis document discusses the following:\n\n- The benefits of using Active Assist in an organization.\n- The challenges that organizations might encounter when they adopt Active Assist at enterprise-scale.\n- How to design automation pipelines using Active Assist.\n\nThe series consists of the following parts:\n\n- Patterns for using Active Assist at scale (this document)\n- [Using serverless pipelines with Active Assist](/recommender/docs/using-serverless-pipelines-with-active-assist)\n- [Using the GKE Enterprise toolchain with Active Assist](/recommender/docs/using-anthos-toolchain-with-active-assist)\n\nActive Assist\n-------------\n\nActive Assist is a portfolio of tools that use data, intelligence, and machine\nlearning to reduce cloud complexity and administrative work, helping enterprises\nto optimize the security, performance, manageability, and cost of their cloud.\n\nMany enterprises have a mandate to ensure that [the principle of least privilege](https://wikipedia.org/wiki/Principle_of_least_privilege)\nis applied to their business applications and infrastructure. Enterprises also\nwant to minimize resource waste and maximize the performance of business applications\nwhile also reducing administrative work and cost. As a consequence, IT departments often\nface scrutiny and pressure to meet these requirements with speed and agility.\nActive Assist gives them tools that they can use to help meet these goals.\n\nCloud optimization for enterprises\n----------------------------------\n\nBecause workloads, infrastructure, security needs, and processes are unique to\neach enterprise, you must adapt cloud optimization strategies to meet your specific\nneeds.\n\nIn the context of this document, cloud optimization strategies for your\nGoogle Cloud footprint focus on how you can leverage the Active Assist portfolio\nwhen you plan and design optimization strategies.\n\n### Defining a vision and understanding drivers\n\nIt's important for enterprises to define the issues that they want to use to inform\ntheir approach for cloud footprint optimization. The following are common issues:\n\n- Security\n- Performance\n- Cost optimization\n- Agility\n\n### Enterprise goals\n\nWhen you begin to architect an automation pipeline for Active Assist recommendations,\nyou should start by defining the goals for your enterprise and assigning priorities\nto each objective. You can then map these priorities to a roadmap for rolling out\nand scaling Active Assist in your Google Cloud organization.\n\nFor example, an enterprise might want to use Active Assist recommendations for\nsecurity and cost optimization. However, the enterprise might initially choose\nto invest in building an automation pipeline for the security-related\nrecommendations that Active Assist generates. At a later stage, as the enterprise\ngains more experience in using the Active Assist portfolio and matures in their\nautomation journey, it might automate other types of\n[recommendations](/recommender/docs/recommenders),\nfor example, [VM rightsizing](/compute/docs/instances/apply-sizing-recommendations-for-instances) and\n[Idle VM Recommender](/compute/docs/instances/viewing-and-applying-idle-vm-recommendations).\n\nDesigning a strategy\n--------------------\n\nEnterprises must have a clearly defined process for how they want to review\nand actuate the recommendations that Active Assist generates. We recommend a\nphased approach that incorporates an increasing degree of automation in a\nmeasured manner. An iterative approach that enterprises can take when adopting\nActive Assist in their Google Cloud organization is as follows:\n\n- **Phase one** :\n - Review Active Assist recommendations in the [Google Cloud console](/cloud-console).\n - Export the recommendations to [BigQuery](/recommender/docs/bq-export/export-recommendations-to-bq).\n- **Phase two** :\n - Use Recommender APIs.\n- **Phase three** :\n - Integrate recommendations review into DevOps pipelines.\n\nThis approach lets you iteratively incorporate more automation into your\nActive Assist recommendations pipelines.\n\n### Phase one: Reviewing Active Assist recommendations in Google Cloud console\n\nIn the first phase, you review Active Assist recommendations in the\nGoogle Cloud console using\n[Recommendation Hub](/recommender/docs/recommendation-hub/identify-configuration-problems#:%7E:text=The%20Recommendation%20Hub%20is%20a,these%20in%20a%20central%20location.).\nYou use a console-based approach to review and implement recommendations. This\napproach helps you gain familiarity with Active Assist recommendations while\nassessing their suitability. It also helps you to decide which recommendation\ncategories you want to prioritize. As shown in the following image,\nRecommendation Hub lets you review recommendations for each\nresource category that recommendations are available for and drill into the relevant\ndetails for each resource within the group.\n\nEnterprise teams can export recommendations to BigQuery. Exporting\nrecommendations to BigQuery lets you review recommendations at\nscale across the organization. It also lets you run queries in specific areas\nof interest for your enterprise. You can also consider [building a dashboard](https://cloud.google.com/blog/topics/cost-management/manage-cloud-costs-with-new-oss-recommendations-dashboard) to\nhelp your team better view and manage your recommendations.\n\n### Phase two: Using the Recommender APIs\n\nIn the second phase, you combine automation with manual reviews and validations to\nimplement recommendations generated by Active Assist. This approach helps you to gain\nagility. It also lets you make the most of platform-generated recommendations at scale,\nwhile retaining tight control on how recommendations are implemented.\n\nYou learn how this approach can be realized in\n[Using Serverless pipelines with Active Assist](/architecture/using-serverless-pipelines-with-active-assist).\n\n### Phase three: Integration recommendations into DevOps pipelines\n\nIn the third phase, you bring the review of recommendations into your DevOps\npipeline. You inject recommendations management and analysis into the DevOps\npipeline, enabling a streamlined process for resource and recommendations management.\nThis approach also enables the development of an approvals process that your\nteams might already be using as part of the continuous integration and continuous\ndeployment (CI/CD) process. This step relies more\nheavily on automation and code-based analysis of recommendations than phase two.\n\nBecause this approach needs an initial investment of effort to develop the automation\nframework, we recommend that you don't implement this phase until you have a well-established\nDevOps strategy.\n\nYou can learn about how this approach works in the following tutorial:\n\n- [Using recommendations for Infrastructure as Code](/recommender/docs/tutorial-iac)\n\nWhen you have a defined strategy for the adoption of Active Assist, the next step\nis to execute and roll out your phased approach.\n\nWhat's next\n-----------\n\n- Learn about how to [use recommendations for Infrastructure as Code](/recommender/docs/tutorial-iac).\n- Read how Active Assist can [help you to optimize Google Cloud resources](https://cloud.google.com/blog/products/management-tools/optimize-google-cloud-resources-with-active-assist).\n- Learn about [modern CI/CD with GKE](/kubernetes-engine/docs/tutorials/modern-cicd-gke-user-guide).\n- See how you can [achieve least privilege access using Policy Intelligence](https://www.youtube.com/watch?v=LYUVnvRovIM&feature=youtu.be).\n- Read about [using IAM Recommender to bulk-apply least privilege principles](https://cloud.google.com/blog/products/identity-security/using-iam-recommender-to-bulk-apply-least-privilege-principles)."]]