AWS emr documentation change
Summary
Minor grammatical and phrasing improvements (e.g., 'you must meet' to 'meet', 'see' to 'refer to', 'recommend' to 'suggest'). No technical changes to permissions, security controls, or features.
Security assessment
Changes are purely editorial/grammatical. No modifications to security-related content like permissions, roles, or access controls. References to security documentation (e.g., IAM permissions) remain unchanged except for phrasing.
Diff
diff --git a/emr/latest/EMR-Serverless-UserGuide/interactive-workloads.md b/emr/latest/EMR-Serverless-UserGuide/interactive-workloads.md index b557ec203..3ec93f42f 100644 --- a//emr/latest/EMR-Serverless-UserGuide/interactive-workloads.md +++ b//emr/latest/EMR-Serverless-UserGuide/interactive-workloads.md @@ -9 +9 @@ OverviewPrerequisitesPermissionsConfigurationConsiderations -With EMR Serverless interactive applications, you can run interactive workloads for Spark with EMR Serverless using notebooks that are hosted in EMR Studio. +With EMR Serverless interactive applications, run interactive workloads for Spark with EMR Serverless using notebooks that are hosted in EMR Studio. @@ -28 +28 @@ Use cases for interactive applications in EMR Serverless include the following: -To use interactive workloads with EMR Serverless, you must meet the following requirements: +To use interactive workloads with EMR Serverless, meet the following requirements: @@ -32 +32 @@ To use interactive workloads with EMR Serverless, you must meet the following re - * To access your interactive application, execute the workloads that you submit, and run interactive notebooks from EMR Studio, you need specific permissions and roles. For more information, see Required permissions for interactive workloads. + * To access your interactive application, execute the workloads that you submit, and run interactive notebooks from EMR Studio, you need specific permissions and roles. For more information, refer to Required permissions for interactive workloads. @@ -39 +39 @@ To use interactive workloads with EMR Serverless, you must meet the following re -In addition to the basic [permissions that are required to access EMR Serverless](./setting-up.html#setting-up-iam), you must configure additional permissions for your IAM identity or role: +In addition to the basic [permissions that are required to access EMR Serverless](./setting-up.html#setting-up-iam), configure additional permissions for your IAM identity or role: @@ -44 +44 @@ In addition to the basic [permissions that are required to access EMR Serverless -Set up user and Workspace permissions for EMR Studio. For more information, see [Configure EMR Studio user permissions](https://docs.aws.amazon.com/emr/latest/ManagementGuide/emr-studio-user-permissions.html) in the _Amazon EMR Management Guide_. +Set up user and Workspace permissions for EMR Studio. For more information, refer to [Configure EMR Studio user permissions](https://docs.aws.amazon.com/emr/latest/ManagementGuide/emr-studio-user-permissions.html) in the _Amazon EMR Management Guide_. @@ -49 +49 @@ Set up user and Workspace permissions for EMR Studio. For more information, see -Set up a job runtime role. For more information, see [Create a job runtime role](./getting-started.html#gs-runtime-role). +Set up a job runtime role. For more information, refer to [Create a job runtime role](./getting-started.html#gs-runtime-role). @@ -108 +108 @@ Use the following high-level steps to create an EMR Serverless application with - 2. Then, launch a workspace from EMR Studio and attach to an EMR Serverless application as a compute option. For more information, see the **Interactive workload** tab in Step 2 of the [EMR Serverless Getting Started](https://docs.aws.amazon.com/emr/latest/EMR-Serverless-UserGuide/getting-started.html#gs-job-run-console) documentation. + 2. Then, launch a workspace from EMR Studio and attach to an EMR Serverless application as a compute option. For more information, refer to the **Interactive workload** tab in Step 2 of the [EMR Serverless Getting Started](https://docs.aws.amazon.com/emr/latest/EMR-Serverless-UserGuide/getting-started.html#gs-job-run-console) documentation. @@ -129 +129 @@ When you attach an application to a Studio Workspace, the application start trig - * For an optimized startup experience, we recommend that you configure pre-initialized capacity for drivers and executors, and that you pre-start your application. When you pre-start the application, you ensure that it's ready when you want to attach it to your Workspace. + * For an optimized startup experience, we suggest that you configure pre-initialized capacity for drivers and executors, and that you pre-start your application. When you pre-start the application, you ensure that it's ready when you want to attach it to your Workspace. @@ -136 +136 @@ When you attach an application to a Studio Workspace, the application start trig - * When using an interactive application, we recommend that you configure a pre-intialized capacity of kernels, drivers, and executors to run your notebooks. Each Spark interactive session requires one kernel and one driver, so EMR Serverless maintains a pre-initialized kernel worker for every pre-initialized driver. By default, EMR Serverless maintains a pre-initialized capacity of one kernel worker throughout the entire application even if you don't specify any pre-initialized capacity for drivers. Each kernel worker uses 4 vCPU and 16 GB of memory. For current pricing information, see the [Amazon EMR Pricing](https://aws.amazon.com/emr/pricing/) page. + * When using an interactive application, we suggest that you configure a pre-intialized capacity of kernels, drivers, and executors to run your notebooks. Each Spark interactive session requires one kernel and one driver, so EMR Serverless maintains a pre-initialized kernel worker for every pre-initialized driver. By default, EMR Serverless maintains a pre-initialized capacity of one kernel worker throughout the entire application even if you don't specify any pre-initialized capacity for drivers. Each kernel worker uses 4 vCPU and 16 GB of memory. For current pricing information, refer to the [Amazon EMR Pricing](https://aws.amazon.com/emr/pricing/) page. @@ -138 +138 @@ When you attach an application to a Studio Workspace, the application start trig - * You must have sufficient vCPU service quota in your AWS account to run interactive workloads. If you don't run Lake Formation-enabled workloads, we recommend at least 24 vCPU. If you do, we recommend at least 28 vCPU. + * You must have sufficient vCPU service quota in your AWS account to run interactive workloads. If you don't run Lake Formation-enabled workloads, we suggest at least 24 vCPU. If you do, we suggest at least 28 vCPU. @@ -142 +142 @@ When you attach an application to a Studio Workspace, the application start trig - * To enable Lake Formation with interactive workloads, set the configuration `spark.emr-serverless.lakeformation.enabled` to `true` under the `spark-defaults` classification in the `runtime-configuration` object when you [create an EMR Serverless application](https://docs.aws.amazon.com/emr/latest/EMR-Serverless-UserGuide/getting-started.html). To learn more about enabling Lake Formation in EMR Serverless, see [Enabling Lake Formation in Amazon EMR](https://docs.aws.amazon.com/emr/latest/EMR-Serverless-UserGuide/emr-serverless-lf-enable.html#emr-serverless-lf-enable-config). + * To enable Lake Formation with interactive workloads, set the configuration `spark.emr-serverless.lakeformation.enabled` to `true` under the `spark-defaults` classification in the `runtime-configuration` object when you [create an EMR Serverless application](https://docs.aws.amazon.com/emr/latest/EMR-Serverless-UserGuide/getting-started.html). To learn more, refer to [Enabling Lake Formation in Amazon EMR](https://docs.aws.amazon.com/emr/latest/EMR-Serverless-UserGuide/emr-serverless-lf-enable.html#emr-serverless-lf-enable-config).