使用 Azure 入口網站來管理 Azure Data Lake AnalyticsManage Azure Data Lake Analytics using the Azure portal

本文說明如何使用 Azure 入口網站來管理 Azure Data Lake Analytics 帳戶、資料來源、使用者和作業。This article describes how to manage Azure Data Lake Analytics accounts, data sources, users, and jobs by using the Azure portal.

管理 Data Lake Analytics 帳戶Manage Data Lake Analytics accounts

建立帳戶Create an account

  1. 登入 Azure 入口網站Sign in to the Azure portal.
  2. 按一下 [建立資源] > [智慧 + 分析] > [Data Lake Analytics] 。Click Create a resource > Intelligence + analytics > Data Lake Analytics.
  3. 選取下列項目的值︰Select values for the following items:
    1. 名稱:Data Lake Analytics 帳戶的名稱。Name: The name of the Data Lake Analytics account.
    2. 訂用帳戶 :用於此帳戶的 Azure 訂用帳戶。Subscription: The Azure subscription used for the account.
    3. 資源群組:要在其中建立帳戶的 Azure 資源群組。Resource Group: The Azure resource group in which to create the account.
    4. 位置:Data Lake Analytics 帳戶的 Azure 資料中心。Location: The Azure datacenter for the Data Lake Analytics account.
    5. Data Lake Store:Data Lake Analytics 帳戶所要使用的預設存放區。Data Lake Store: The default store to be used for the Data Lake Analytics account. Azure Data Lake Store 帳戶和 Data Lake Analytics 帳戶必須位於相同位置。The Azure Data Lake Store account and the Data Lake Analytics account must be in the same location.
  4. 按一下 [建立] 。Click Create.

刪除 Data Lake Analytics 帳戶Delete a Data Lake Analytics account

在您刪除 Data Lake Analytics 帳戶前,請先刪除其預設 Data Lake Store 帳戶。Before you delete a Data Lake Analytics account, delete its default Data Lake Store account.

  1. 在 Azure 入口網站中,移至您的 Data Lake Analytics 帳戶。In the Azure portal, go to your Data Lake Analytics account.
  2. 按一下 [刪除] 。Click Delete.
  3. 輸入帳戶名稱。Type the account name.
  4. 按一下 [刪除] 。Click Delete.

管理資料來源Manage data sources

Data Lake Analytics 支援下列資料來源:Data Lake Analytics supports the following data sources:

  • Data Lake StoreData Lake Store
  • Azure 儲存體Azure Storage

您可以使用 [資料總管] 來瀏覽資料來源和執行基本檔案管理作業。You can use Data Explorer to browse data sources and perform basic file management operations.

建立資料來源Add a data source

  1. 在 Azure 入口網站中,移至您的 Data Lake Analytics 帳戶。In the Azure portal, go to your Data Lake Analytics account.

  2. 按一下 [資料來源] 。Click Data Sources.

  3. 按一下 [ 新增資料來源]。Click Add Data Source.

    • 若要新增 Azure Data Lake Store 帳戶,您需要帳戶名稱及帳戶的存取權才可進行查詢。To add a Data Lake Store account, you need the account name and access to the account to be able to query it.
    • 若要新增 Azure Blob 儲存體,您需要儲存體帳戶和帳戶金鑰。To add Azure Blob storage, you need the storage account and the account key. 若要找到它們,請在入口網站中移至此儲存體帳戶。To find them, go to the storage account in the portal.

設定防火牆規則Set up firewall rules

您可以使用 Data Lake Analytics,進一步在網路層級鎖定 Data Lake Analytics 帳戶的存取。You can use Data Lake Analytics to further lock down access to your Data Lake Analytics account at the network level. 您可以啟用防火牆、指定 IP 位址或為受信任的用戶端定義 IP 位址範圍。You can enable a firewall, specify an IP address, or define an IP address range for your trusted clients. 啟用這些量值之後,只有具有定義範圍內 IP 位址的用戶端可以連線到存放區。After you enable these measures, only clients that have the IP addresses within the defined range can connect to the store.

如果有其他 Azure 服務 (例如 Azure Data Factory 或 VM) 連線到 Data Lake Analytics 帳戶,請確定 [允許 Azure 服務] 已 [開啟] 。If other Azure services, like Azure Data Factory or VMs, connect to the Data Lake Analytics account, make sure that Allow Azure Services is turned On.

設定防火牆規則Set up a firewall rule

  1. 在 Azure 入口網站中,移至您的 Data Lake Analytics 帳戶。In the Azure portal, go to your Data Lake Analytics account.
  2. 在左側功能表上,按一下 [防火牆] 。On the menu on the left, click Firewall.

新增使用者Add a new user

您可以使用 [新增使用者精靈] 輕鬆地佈建新的 Data Lake 使用者。You can use the Add User Wizard to easily provision new Data Lake users.

  1. 在 Azure 入口網站中,移至您的 Data Lake Analytics 帳戶。In the Azure portal, go to your Data Lake Analytics account.
  2. 在左側 [快速入門] 之下,按一下 [新增使用者精靈] 。On the left, under Getting Started, click Add User Wizard.
  3. 選取使用者,然後按一下 [選取] 。Select a user, and then click Select.
  4. 選取角色,然後按一下 [選取] 。Select a role, and then click Select. 若要設定新的開發人員以使用 Azure Data Lake,請選取 [Data Lake Analytics 開發人員] 角色。To set up a new developer to use Azure Data Lake, select the Data Lake Analytics Developer role.
  5. 選取 U-SQL 資料庫的存取控制清單 (ACL)。Select the access control lists (ACLs) for the U-SQL databases. 當您對您的選擇感到滿意時,請按一下 [選取] 。When you're satisfied with your choices, click Select.
  6. 選取檔案的 ACL。Select the ACLs for files. 對於預設存放區,請不要變更根資料夾 "/" 和 /system 資料夾的 ACL。For the default store, don't change the ACLs for the root folder "/" and for the /system folder. 按一下 [選取] 。Click Select.
  7. 檢閱您選取的所有變更,然後按一下 [執行] 。Review all your selected changes, and then click Run.
  8. 當精靈完成時,按一下 [完成] 。When the wizard is finished, click Done.

管理角色型存取控制Manage Role-Based Access Control

如同其他 Azure 服務,您可以使用角色型存取控制 (RBAC) 來控制使用者與服務互動的方式。Like other Azure services, you can use Role-Based Access Control (RBAC) to control how users interact with the service.

標準 RBAC 角色具有下列功能:The standard RBAC roles have the following capabilities:

  • 擁有者:可以提交、監視、取消任何使用者的作業,以及設定帳戶。Owner: Can submit jobs, monitor jobs, cancel jobs from any user, and configure the account.
  • 參與者:可以提交、監視、取消任何使用者的作業,以及設定帳戶。Contributor: Can submit jobs, monitor jobs, cancel jobs from any user, and configure the account.
  • 讀取者:可以監視作業。Reader: Can monitor jobs.

使用 Data Lake Analytics 開發人員角色來啟用 U-SQL 開發人員,以使用 Data Lake Analytics 服務。Use the Data Lake Analytics Developer role to enable U-SQL developers to use the Data Lake Analytics service. 您可以使用 Data Lake Analytics 開發人員角色來:You can use the Data Lake Analytics Developer role to:

  • 提交作業。Submit jobs.
  • 監視任何使用者所提交作業的作業狀態和進度。Monitor job status and the progress of jobs submitted by any user.
  • 請參閱任何使用者所提交作業中的 U-SQL 指令碼。See the U-SQL scripts from jobs submitted by any user.
  • 只取消您自己的作業。Cancel only your own jobs.

將使用者或安全性群組新增到 Data Lake Analytics 帳戶Add users or security groups to a Data Lake Analytics account

  1. 在 Azure 入口網站中,移至您的 Data Lake Analytics 帳戶。In the Azure portal, go to your Data Lake Analytics account.
  2. 按一下 [存取控制 (IAM)] > [新增角色指派] 。Click Access control (IAM) > Add role assignment.
  3. 選取角色。Select a role.
  4. 新增使用者。Add a user.
  5. 按一下 [確定] 。Click OK.

注意

如果使用者或安全性群組需要提交作業,他們也需要有存放區帳戶的權限。If a user or a security group needs to submit jobs, they also need permission on the store account. 如需詳細資訊,請參閱保護儲存在 Data Lake Store 中的資料For more information, see Secure data stored in Data Lake Store.

管理工作Manage jobs

提交作業Submit a job

  1. 在 Azure 入口網站中,移至您的 Data Lake Analytics 帳戶。In the Azure portal, go to your Data Lake Analytics account.

  2. 按一下 [ 新增工作]。Click New Job. 對於每項作業,請設定:For each job, configure:

    1. 作業名稱:作業的名稱。Job Name: The name of the job.
    2. 優先順序:數字越小,優先順序越高。Priority: Lower numbers have higher priority. 如果有兩項作業排入佇列,優先順序值較小的作業會優先執行。If two jobs are queued, the one with lower priority value runs first.
    3. 平行處理原則:要為此作業保留的計算程序數目上限。Parallelism: The maximum number of compute processes to reserve for this job.
  3. 按一下 [ 提交作業]。Click Submit Job.

監視工作Monitor jobs

  1. 在 Azure 入口網站中,移至您的 Data Lake Analytics 帳戶。In the Azure portal, go to your Data Lake Analytics account.
  2. 按一下 [檢視所有作業] 。Click View All Jobs. 隨即顯示帳戶中所有作用中和最近完成的作業清單。A list of all the active and recently finished jobs in the account is shown.
  3. 選擇性,按一下 [篩選] 可協助您依照 [時間範圍] 、[作業名稱] 和 [作者] 值尋找作業。Optionally, click Filter to help you find the jobs by Time Range, Job Name, and Author values.

監視管線作業Monitoring pipeline jobs

屬於某個管線的作業會搭配運作 (通常會循序進行),以完成特定案例。Jobs that are part of a pipeline work together, usually sequentially, to accomplish a specific scenario. 例如,您可以有一個管線,來清除、擷取、轉換、彙總客戶深入解析的使用。For example, you can have a pipeline that cleans, extracts, transforms, aggregates usage for customer insights. 提交作業之後,可使用 [管線] 屬性來找到管線作業。Pipeline jobs are identified using the "Pipeline" property when the job was submitted. 使用 ADF V2 排程的作業會自動填入此屬性。Jobs scheduled using ADF V2 will automatically have this property populated.

若要檢視屬於管線的 U-SQL 作業清單:To view a list of U-SQL jobs that are part of pipelines:

  1. 在 Azure 入口網站中,移至您的 Data Lake Analytics 帳戶。In the Azure portal, go to your Data Lake Analytics accounts.
  2. 按一下 [作業深入解析] 。Click Job Insights. 預設會開啟 [所有作業] 索引標籤,其中顯示執行中、已佇列和已結束的作業清單。The "All Jobs" tab will be defaulted, showing a list of running, queued, and ended jobs.
  3. 按一下 [管線作業] 索引標籤。這會顯示管線作業清單,以及每個管線的彙總統計資料。Click the Pipeline Jobs tab. A list of pipeline jobs will be shown along with aggregated statistics for each pipeline.

監視週期性作業Monitoring recurring jobs

週期性作業是具有相同商務邏輯,但每次執行都會使用不同輸入資料的作業。A recurring job is one that has the same business logic but uses different input data every time it runs. 在理想情況下,週期性作業一律會成功,而且執行時間相當穩定;監視這些行為有助於確保作業狀況良好。Ideally, recurring jobs should always succeed, and have relatively stable execution time; monitoring these behaviors will help ensure the job is healthy. 可使用 [週期性] 屬性來找到週期性作業。Recurring jobs are identified using the "Recurrence" property. 使用 ADF V2 排程的作業會自動填入此屬性。Jobs scheduled using ADF V2 will automatically have this property populated.

若要檢視週期性的 U-SQL 作業清單:To view a list of U-SQL jobs that are recurring:

  1. 在 Azure 入口網站中,移至您的 Data Lake Analytics 帳戶。In the Azure portal, go to your Data Lake Analytics accounts.
  2. 按一下 [作業深入解析] 。Click Job Insights. 預設會開啟 [所有作業] 索引標籤,其中顯示執行中、已佇列和已結束的作業清單。The "All Jobs" tab will be defaulted, showing a list of running, queued, and ended jobs.
  3. 按一下 [週期性作業] 索引標籤。這會顯示週期性作業清單,以及每個週期性作業的彙總統計資料。Click the Recurring Jobs tab. A list of recurring jobs will be shown along with aggregated statistics for each recurring job.

後續步驟Next steps