AWSCertifiedBigDataSlides-compressed

Exam Overview

• Course: AWS Certified Data Analytics Specialty Course DAS-C01

• Objective: Prepare for the Big Data Specialty certification exam (BDS-C00).

• Recommended Background:

  • Prior knowledge of AWS services (EC2, networking).

  • Familiarity with data and analytics concepts.

• Duration: Long and interesting course, take your time.

Instructors

• Stephane Maarek

  • IT Consultant & AWS Big Data Architect.

  • Veteran instructor; 94% certification score.

  • Links: GitHub, LinkedIn, Medium, Twitter.

• Frank Kane

  • Former Amazon Sr. Software Engineer and Manager.

  • Focus on Machine Learning and Big Data.

  • Owner of Sundog Education.

  • Links: LinkedIn, Twitter, Facebook.

Course Coverage

• AWS Big Data Services:

  • Amazon Kinesis, AWS Lambda, AWS Glue, Amazon EMR, Amazon ML, SageMaker, etc.

  • Services categorized into:

    • Collection: Kinesis, AWS IoT, SQS.

    • Storage: S3, DynamoDB.

    • Processing: AWS Lambda, Glue, EMR.

    • Analysis: Athena, Redshift, QuickSight.

    • Security: AWS KMS, CloudHSM.

Case Study

• Case Study Overview: cadabra.com

• Requirements:

  • Order History App: Client app, server logs.

  • Product Recommendations: Server logs.

  • Transaction Rate Alarm: Server logs.

  • Near Real-Time Log Analysis: Amazon OpenSearch, Kinesis Data Firehose.

  • Data Warehousing & Visualization: (Managed Serverless).

AWS Data Collection Methods

• Real-Time:

  • Tools: Kinesis Data Streams, SQS.

• Near Real-Time:

  • Tools: Kinesis Data Firehose, DMS.

• Batch Analysis:

  • Tools: Snowball, Data Pipeline.

Kinesis Data Streams

• Architecture: Consists of shards, producers, consumers.

• Data Ingestion Constraints:

  • Retention (1 to 365 days).

  • Record limitations (1 MB/record, 1,000 records/sec per shard).

• Security:

  • Control access using IAM,

  • Encryption in flight and at rest with KMS.

AWS Lambda

• Functionality: Serverless computing and event-driven architecture.

• Common Integrations:

  • Kinesis, S3, DynamoDB.

• Cost Model: Pay for number of requests and duration of execution.

• Supported Languages: Node.js, Python, Java, C#, Go.

AWS Glue

• Serverless ETL Service: Effective for data cleaning and transformation.

• Crawler Feature: Automatic schema discovery from data sources.

• Key Transformations:

  • Machine learning transformations (FindMatches).

  • Supports a variety of data formats.

Amazon Redshift

• Service Type: Fully-managed, petabyte-scale data warehouse.

• Performance: 10x faster than others via MPP and columnar storage.

• Key Features:

  • Scaling, backups, and concurrency management.

  • Storage options: RA3 nodes offering independent scaling of compute and storage.

Amazon QuickSight

• Functionality: Fast, serverless business analytics service.

• Key Features: Commonly used for dashboards, ad-hoc analysis, and visualizations.

• Security: Multi-factor authentication, IAM policies, row-level security.

Exam Preparation Tips

• Timing: 65 questions in 170 minutes (~2.5 minutes per question).

• Practice: Take practice exams; familiarize with AWS white papers and make use of the AWS training resource.

• Day of Exam:

  • Bring two forms of ID; no notes or electronic devices allowed.

  • Arrive early and prepared to reduce stress.

Additional Resources

• Review AWS Big Data White Paper and specific service documentation for exam preparation.

• Join AWS forums and communities for tips and advice from others who have taken the exam.