AWS lambda documentation change
Summary
Updated terminology from 'ESM' to 'event source mapping' and 'BatchingWindowInSeconds' to 'MaximumBatchingWindowInSeconds'. Added recommendations for provisioned concurrency/SnapStart and clarified low latency configuration details.
Security assessment
Changes focus on terminology standardization and performance optimization guidance. No security vulnerabilities or security-specific features are mentioned. Recommendations about cold starts and concurrency relate to performance rather than security.
Diff
diff --git a/lambda/latest/dg/with-kafka-low-latency.md b/lambda/latest/dg/with-kafka-low-latency.md index e868e4420..c607c7b31 100644 --- a//lambda/latest/dg/with-kafka-low-latency.md +++ b//lambda/latest/dg/with-kafka-low-latency.md @@ -13 +13 @@ AWS Lambda natively supports low latency event processing for applications that -To enable low latency processing on a Kafka event source mapping (ESM), the following basic configuration is required: +To enable low latency processing on a Kafka event source mapping, the following basic configuration is required: @@ -15 +15 @@ To enable low latency processing on a Kafka event source mapping (ESM), the foll - * Enable Provisioned mode. For more information, see [Configuring provisioned mode](./with-kafka-process.html#services-kafka-provisioned-mode). + * Enable provisioned mode. For more information, see [Configuring provisioned mode](./with-kafka-process.html#services-kafka-provisioned-mode). @@ -17 +17 @@ To enable low latency processing on a Kafka event source mapping (ESM), the foll - * Set the ESM's `BatchingWindowInSeconds` parameter to 0. For more information, see [Batching behavior](./invocation-eventsourcemapping.html#invocation-eventsourcemapping-batching). + * Set the event source mapping's `MaximumBatchingWindowInSeconds` parameter to 0. For more information, see [Batching behavior](./invocation-eventsourcemapping.html#invocation-eventsourcemapping-batching). @@ -24 +24 @@ To enable low latency processing on a Kafka event source mapping (ESM), the foll -Consider the following recommendations to optimize your Kafka ESM for low latency: +Consider the following recommendations to optimize your Kafka event source mapping for low latency: @@ -28 +28 @@ Consider the following recommendations to optimize your Kafka ESM for low latenc -In Provisioned mode for Kafka ESMs, Lambda allows you to fine-tune the throughput of your ESM by configuring a minimum and maximum number of resources called **event pollers**. An event poller (or **a poller**) represents a compute resource that underpins an ESM in the provisioned mode, and allocates up to 5 MB/s throughput. Each event poller supports up to 5 concurrent Lambda invocations. +In provisioned mode for Kafka event source mapping, Lambda allows you to fine-tune the throughput of your event source mapping by configuring a minimum and maximum number of resources called **event pollers**. An event poller (or **a poller**) represents a compute resource that underpins an event source mapping in the provisioned mode, and allocates up to 5 MB/s throughput. Each event poller supports up to 5 concurrent Lambda invocations. @@ -36 +36 @@ With a batch size of 20 records and an average target function duration of 50ms, -It's recommended to provision additional event pollers as buffer to avoid consistently operating at maximum capacity. +We recommend to provision additional event pollers as buffer to avoid consistently operating at maximum capacity. @@ -44 +44 @@ Some of the additional considerations include: - * Cold starts from the invocation of your Lambda target function can potentially increase end-to-end latency. To reduce this risk, consider enabling [provisioned concurrency](./provisioned-concurrency.html) or [SnapStart](./snapstart.html) on your ESM's target function. Additionally, optimize your function's memory allocation to ensure consistent and optimal executions. + * Cold starts from the invocation of your Lambda target function can potentially increase end-to-end latency. To reduce this risk, consider enabling [provisioned concurrency](./provisioned-concurrency.html) or [SnapStart](./snapstart.html) on your event source mapping's target function. Additionally, optimize your function's memory allocation to ensure consistent and optimal executions. @@ -46 +46 @@ Some of the additional considerations include: - * When `BatchingWindowInSeconds` is set to 0, Lambda will immediately process any available records without waiting to fill the complete batch size. For example, if your batch size is set to 1,000 records but only 100 records are available, Lambda will process those 100 records immediately rather than waiting for the full 1,000 records to accumulate. + * When `MaximumBatchingWindowInSeconds` is set to 0, Lambda will immediately process any available records without waiting to fill the complete batch size. For example, if your batch size is set to 1,000 records but only 100 records are available, Lambda will process those 100 records immediately rather than waiting for the full 1,000 records to accumulate.