A company is migrating its existing on-premises ETL jobs to Amazon EMR. The code consists of a series of jobs written in Java. The company needs to reduce overhead for the system administrators without changing the underlying code. Due to the sensitivity of the data, compliance requires that the company use root device volume encryption on all nodes in the cluster. Corporate standards require that environments be provisioned though AWS CloudFormation when possible.Which solution satis es these requirements?
Answer(s): C
Enabling encryption at the AMI level ensures root device volumes are encrypted on all EMR nodes by default, meeting the compliance requirement without per-node bootstrap steps and without code changes. Using a custom AMI preserves existing Java ETL logic while aligning with CloudFormation provisioning for EMR.A) Incorrect: Installing open-source Hadoop on EC2 bypasses managed EMR benefits and requires manual maintenance; not aligned with EMR-provisioning best practices and encryption per AWS-managed root volumes. B) Incorrect: TLS bootstrap does not encrypt root volumes; it secures in-transit data but not disk at rest. D) Incorrect: A bootstrap action to encrypt root volumes is not supported; root device encryption is configured via the AMI or EBS volume settings, not via bootstrap.
https://docs.aws.amazon.com/emr/latest/ManagementGuide/emr-custom-ami.html
A transportation company uses IoT sensors attached to trucks to collect vehicle data for its global delivery eet. The company currently sends the sensor data in small .csv les to Amazon S3. The les are then loaded into a 10-node Amazon Redshift cluster with two slices per node and queried using both Amazon Athena and Amazon Redshift. The company wants to optimize the les to reduce the cost of querying and also improve the speed of data loading into the AmazonRedshift cluster.Which solution meets these requirements?
Answer(s): D
Converting CSV to Parquet and creating multiple Parquet files aligns with columnar storage, which reduces I/O and speeds up queries with Athena and Redshift Spectrum-like workflows, while also optimizing load parallelism in Redshift COPY. Parquet is efficient for nested schemas and supports predicate pushdown, minimizing scanned data.A) Parquet single large file reduces parallelism during COPY and queries; multiple small files can improve load and query performance in distributed systems.B) Avro is row-based and not as storage-efficient as Parquet for analytics workloads; also adds processing steps without clear load/performance benefits over Parquet.C) ORC is a valid columnar format, but combining with Parquet and multi-file strategy is not as widely supported in multi-engine workflows as Parquet; single large ORC file may hinder parallelism.D) Parquet files (20 files) maximize parallel LOAD and scans, optimizing both Redshift COPY throughput and Athena queries.
An online retail company with millions of users around the globe wants to improve its ecommerce analytics capabilities. Currently, clickstream data is uploaded directly to Amazon S3 as compressed les. Several times each day, an application running on Amazon EC2 processes the data and makes search options and reports available for visualization by editors and marketers. The company wants to make website clicks and aggregated data available to editors and marketers in minutes to enable them to connect with users more effectively. Which options will help meet these requirements in the MOST e cient way? (Choose two.)
Answer(s): A,D
The combination of Kinesis Data Firehose delivering near-real-time to OpenSearch Service and OpenSearch Dashboards for live visualization provides minimal latency end-to-end and scalable analytics.A) Kinesis Data Firehose delivers compressed clickstream records directly to OpenSearch Service with near-real-time ingestion, enabling fast dashboards. This is the most efficient pipeline for streaming data to search analytics.B) Lambda from S3 introduces latency and batch-oriented processing; it’s not as streaming-oriented as Firehose for near-real-time dashboards.C) Self-managed OpenSearch on EC2 adds operational burden and is not as streamlined for near-real-time visualization.D) OpenSearch Dashboards provides real-time filtering, aggregation, and visualization atop OpenSearch Service, enabling near-real-time dashboards.E) Kinesis Data Streams with a consumer adds complexity and potential increased latency compared with Firehose’s managed pipeline.A) Firehose to OpenSearch ServiceD) OpenSearch Dashboards visualizations
A company is streaming its high-volume billing data (100 MBps) to Amazon Kinesis Data Streams. A data analyst partitioned the data on account_id to ensure that all records belonging to an account go to the same Kinesis shard and order is maintained. While building a custom consumer using the Kinesis Java SDK, the data analyst notices that, sometimes, the messages arrive out of order for account_id. Upon further investigation, the data analyst discovers the messages that are out of order seem to be arriving from different shards for the same account_id and are seen when a stream resize runs.What is an explanation for this behavior and what is the solution?
A single sentence summary: After a stream resize, child shards can appear, and a consumer must ensure complete processing of parent shards before consuming from child shards to preserve per-account ordering.A) Incorrect: Shard-level ordering is preserved within a shard; stream resizes create new child shards. The issue is cross-shard ordering during resize, not simply having multiple shards.B) Incorrect: Explicit hash keys on producer side control shard assignment, but the problem described is due to parent/child shard processing order during resize, not hash key distribution.C) Incorrect: PutRecords vs PutRecord affects throughput and per-call sequencing, but per-account ordering across shards during resize is not resolved by switching APIs.D) Correct: During a resize, a parent shard splits into child shards; processing must complete on the parent before consuming children to maintain account_id ordering.
A media analytics company consumes a stream of social media posts. The posts are sent to an Amazon Kinesis data stream partitioned on user_id. An AWSLambda function retrieves the records and validates the content before loading the posts into an Amazon OpenSearch Service (Amazon Elasticsearch Service) cluster. The validation process needs to receive the posts for a given user in the order they were received by the Kinesis data stream.During peak hours, the social media posts take more than an hour to appear in the Amazon OpenSearch Service (Amazon ES) cluster. A data analytics specialist must implement a solution that reduces this latency with the least possible operational overhead.Which solution meets these requirements?
For real-time processing of streaming data, Amazon Kinesis partitions data in multiple shards that can then be consumed by multiple Amazon EC2
https://d1.awsstatic.com/whitepapers/AWS_Cloud_Best_Practices.pdf
A company launched a service that produces millions of messages every day and uses Amazon Kinesis Data Streams as the streaming service. The company uses the Kinesis SDK to write data to Kinesis Data Streams. A few months after launch, a data analyst found that write performance is signi cantly reduced. The data analyst investigated the metrics and determined that Kinesis is throttling the write requests. The data analyst wants to address this issue without signi cant changes to the architecture. Which actions should the data analyst take to resolve this issue? (Choose two.)
Answer(s): C,D
The correct options address shard throughput and key distribution, which prevent throttling without major architecture changes.C) Increasing the number of shards via UpdateShardCount expands total write throughput capacity, alleviating per-shard throttling when traffic is high.D) Choosing partition keys that yield a uniform distribution across shards prevents hot shards, improving aggregate write efficiency and reducing throttling.A) Increasing retention period does not affect per-shard throughput or throttling; it only retains data longer.B) Kinesis Agent is a valid ingestion option but not a targeted fix for throttling due to shard saturation and adds unnecessary change.E) Retry logic helps resilience but does not resolve underlying shard-level throttling; it may mask throttling symptoms rather than fix capacity.
A smart home automation company must e ciently ingest and process messages from various connected devices and sensors. The majority of these messages are comprised of a large number of small les. These messages are ingested using Amazon Kinesis Data Streams and sent to Amazon S3 using a Kinesis data stream consumer application. The Amazon S3 message data is then passed through a processing pipeline built on Amazon EMR running scheduled PySpark jobs.The data platform team manages data processing and is concerned about the e ciency and cost of downstream data processing. They want to continue to usePySpark.Which solution improves the e ciency of the data processing jobs and is well architected?
By migrating to AWS Glue Python jobs that merge small files in S3 and convert to Parquet, the solution reduces small-file inefficiency and enables optimized Spark-based processing with cost-effective, managed ETL tailored for PySpark workloads.A) Incorrect: Kinesis Data Firehose with Parquet helps storage format but EMR PySpark still processes on EMR; no shift to a managed PySpark-optimized path.B) Incorrect: Lambda handles micro-batches or individual records poorly for high-volume streaming; not cost-efficient for large-scale PySpark workloads.C) Incorrect: Redshift is not ideal for near-real-time PySpark processing and adds data movement overhead; breaks the PySpark-on-EMR lineage and efficiency goal.D) Correct: Glue Python jobs enable scalable, serverless ETL, optimize small-file handling, and output Parquet, aligning with PySpark workflows.
A large nancial company is running its ETL process. Part of this process is to move data from Amazon S3 into an Amazon Redshift cluster. The company wants to use the most cost-e cient method to load the dataset into Amazon Redshift.Which combination of steps would meet these requirements? (Choose two.)
Answer(s): A,C
The COPY command with a manifest minimizes data transfer and ensures efficient, parallel loading from S3 into Redshift, providing cost-effective, scalable ingestion. Using temporary staging tables helps manage large loads, enables validation, and can reduce write conflicts and retries, contributing to lower overall load time and cost.A) Uses COPY with manifest for optimized, parallel data load from S3.C) Staging tables allow incremental, controlled loading and easier error handling.B) S3DistCp is for Hadoop data movement, not a Redshift load path.D) UNLOAD writes data from Redshift to S3, not for loading into Redshift.E) Redshift Spectrum queries data in S3; it does not load data into Redshift storage.
https://aws.amazon.com/blogs/big-data/top-8-best-practices-for-high-performance-etl-processing-using-amazon-redshift/
Post your Comments and Discuss Amazon DAS-C01 exam prep with other Community members:
Our website is free, but we have to fight against AI bots and content theft. We're sorry for the inconvenience caused by these security measures. You can access the rest of the DAS-C01 content, but please register or login to continue.