This stage of portfolio assessment occurs early in the cloud journey. It's typically performed close to an executive decision to move an existent portfolio of applications to the cloud or during exploratory phases. The discovery acceleration and initial planning stage focuses on the following: 这一资产组合评估阶段发生在云迁移旅程的早期。它通常在做出将现有应用组合迁移至云端的行政决策前后，或在探索阶段进行。发现加速与初步规划阶段侧重于以下内容：
Understanding business drivers理解业务驱动因素
Identifying existent sources of data识别现有数据源
Evaluating the need for automated discovery tooling评估对自动化发现工具的需求
Establishing base models for application prioritization and migration strategy selection建立应用优先级排序和迁移策略选择的基础模型
These activities lead to the following:这些活动将达成以下成果：
A preliminary analysis of the portfolio资产组合的初步分析
Identification of initial migration candidates确定初步的迁移候选对象
The creation of a directional business case for migration创建迁移的方向性业务案例
An outline of initial plans.初步计划大纲。
The key is to understand who the stakeholders of this stage are and what data requirements they have for moving forward. Stakeholders range from board members and CxOs, to business unit leads, senior managers, and application owners.关键在于了解此阶段的利益相关者是谁，以及他们推进工作所需的数据要求。利益相关者范围广泛，包括董事会成员和 CxO、业务部门负责人、高级经理以及应用所有者。
Tip提示
For details and guidance, see the relevant section in the Application portfolio assessment guide for AWS Cloud migration.有关详细信息和指导，请参阅《AWS 云迁移应用组合评估指南》中的相关章节。
High-level objectives and actions高层目标与行动
Determine stakeholders – Who is impacted by the migration? Who are the decision-makers? Who benefits from this migration? Who is responsible for performing the migration?确定利益相关者 —— 谁会受到迁移的影响？谁是决策者？谁能从此次迁移中受益？谁负责执行迁移？
Identify business drivers – What business outcomes and goals are being pursued as a result of migration (for example, business transformation, cost reduction, agility)? This information will be a key determinant of migration strategy and prioritization.确定业务驱动因素 —— 迁移旨在实现哪些业务成果和目标（例如：业务转型、成本削减、敏捷性）？这些信息将成为决定迁移策略和优先级排序的关键因素。
Identify existing sources of data – Identify and document existing sources of data, such as people, tooling, documentation. Assign a level of trust to each source. For example, programmatic or automated sources are more trusted than institutional knowledge or documents. Over time, lower-trust data sources will be replaced with higher-trust data sources.确定现有数据源 —— 识别并记录现有数据源，例如人员、工具和文档。为每个来源分配信任度。例如，程序化或自动化的来源比机构知识或文档更值得信赖。随着时间的推移，低信任度的数据源将被高信任度的数据源所取代。
Build an initial inventory of applications and infrastructure – Identify application names, primary function, criticality, IT environments, high-level compliance and regulatory requirements, and known dependencies. Which IT assets are associated to the applications? Identify product names and versions, historic performance data, known issues, and risks.构建应用和基础设施的初步清单 —— 识别应用名称、主要功能、关键性、IT 环境、高层合规性和监管要求以及已知依赖关系。哪些 IT 资产与这些应用相关联？识别产品名称和版本、历史性能数据、已知问题和风险。
Identify data gaps and discovery needs – Analyze data gaps and evaluate the need to procure automated discovery tooling. Discovery tool investment decisions should be backed by a need to increase overall confidence in data for accurate analysis. To maintain an up-to-date view of the application portfolio, we recommend specialized tooling for workload discovery. 识别数据缺口和发现需求 —— 分析数据缺口，并评估采购自动化发现工具的必要性。发现工具的投资决策应基于提高数据整体可信度以进行准确分析的需求。为保持应用组合视图的实时更新，我们建议使用专门的工具进行工作负载发现。
Deploy discovery tooling – Procure, install, and configure discovery tooling, if applicable, and create a roll-out plan to targeted systems. Two weeks of programmatic data collection to all targeted systems provide enough data to achieve the outcomes of this stage. However, to refine the outputs, ongoing data collection is required in later stages.部署发现工具 —— 采购、安装并配置发现工具（如适用），并为目标系统制定推广计划。对所有目标系统进行两周的程序化数据收集，足以获得该阶段的预期成果。然而，为了优化输出结果，后续阶段仍需持续进行数据收集。
Collect Total Cost of Ownership (TCO) data – Use collected data to produce and update TCO reports and to create a directional business case. For more information, see the Best practices section.收集总拥有成本 (TCO) 数据 —— 使用收集到的数据来生成和更新 TCO 报告，并创建方向性业务案例。有关更多信息，请参阅“最佳实践”部分。
Use application-rationalization models – Establish or adopt base application rationalization models for migration. Include the selection of key application attributes, weighting for prioritization, and initial R type (rehost, replatform, refactor (re-architect), repurchase, retain, retire) based on the 6 Rs decision tree.使用应用合理化模型 —— 建立或采用基础的应用合理化模型进行迁移。包括选择关键应用属性、优先级排序权重，以及基于“6R”决策树的初始 R 类型（重新托管、重新平台化、重构（重新架构）、购买、保留、淘汰）。
Identify initial migration candidates – What applications could be moved now to establish or extend AWS foundations and gain experience? At this stage, the model should prioritize simple workloads with low levels of dependencies (for example, zero-three). 确定初步迁移候选对象 —— 哪些应用可以立即迁移，以建立或扩展 AWS 基础并积累经验？在此阶段，模型应优先考虑依赖关系较低（例如：0-3 个）的简单工作负载。
Plan ongoing assessments – Plan for the next portfolio assessment activities, such as performing a detailed assessment of prioritized workloads.规划后续评估 —— 为接下来的资产组合评估活动制定计划，例如对优先级较高的工作负载进行详细评估。
Plan communications – Establish a communication plan or cadence for your stakeholders and a scope-control mechanism. It is normal for the scope to change as the migration program progresses. Ensure that there is a single source of truth for portfolio data and that changes to the scope are tracked, assessed, and communicated.规划沟通 —— 为利益相关者建立沟通计划或节奏，并设立范围控制机制。随着迁移计划的推进，范围发生变化是正常的。确保资产组合数据有单一事实来源，并对范围变更进行跟踪、评估和沟通。
Outcomes成果
Documented business drivers, outcomes, goals, and technical guiding principles.记录在案的业务驱动因素、成果、目标和技术指导原则。
An initial inventory of applications and infrastructure, and identified data gaps. This is an initial view of the portfolio that will be iterated and refined in further stages.一份应用和基础设施的初步清单，以及已识别的数据缺口。这是资产组合的初步视图，将在后续阶段进行迭代和完善。
A directional business case and estimated cost to migrate.一份方向性业务案例和估算的迁移成本。
A list of three-five applications that are initial migration candidates.一份包含三到五个初步迁移候选应用的项目清单。
A communication plan for portfolio-related activities, milestones, and scope changes.一份针对资产组合相关活动、里程碑和范围变更的沟通计划。
Best practices最佳实践
Focus on the stage objectives and keep analysis at a high level. Adopt a progressive approach to data collection, and do not wait for a complete dataset before moving forward.专注于阶段目标，并将分析保持在高层级。采取渐进式的数据收集方法，不要等待数据集完整后再推进工作。
Avoid analysis paralysis. Focus on identifying the data gaps and driving actions to close those gaps.避免分析瘫痪。专注于识别数据缺口并推动行动以弥补这些缺口。
Procure specialized discovery tooling. Programmatic data, typically obtained through automated tooling, has the higher level of trust and should be preferred before written documentation, static data, and institutional knowledge.采购专门的发现工具。通常通过自动化工具获取的程序化数据具有更高的可信度，应优先于书面文档、静态数据和机构知识。
Work with the identified stakeholders to remove blockers.与确定的利益相关者合作，消除障碍。
Assign a single-threaded leader to the application portfolio assessment workstream.为应用组合评估工作流指派一名单一职责负责人。
Consider using Migration Evaluator
for your directional business case, or explore the AWS Partner Network tools and offerings for TCO and AWS pricing.考虑使用 Migration Evaluator 来制定您的方向性业务案例，或探索 AWS 合作伙伴网络中用于 TCO 和 AWS 定价的工具与服务。 Consider AWS Professional Services
and AWS Partners that could help you accelerate business outcomes.考虑借助 AWS 专业服务和 AWS 合作伙伴来加速实现业务成果。
