Learning AWS Day by Day — Day 43 — Amazon Athena

Exploring AWS !!

Day 43:

Amazon Athena

Serverless, Interactive query Platform, works on data stored in S3.
It uses standard SQL for querying data
It supports variety of formats: CSV, JSON, Avro, ORC (Columnar), Parquet (Columnar)
Only pay for querying the data
Quickly query unstructured, semi-structured and structured data.
Uses Presto (SQL query engine for Big Data)
Ways to access — AWS Console, Athena API, Athena CLI, JDBC connection
Integrates with AWS Glue Data Catalogue
Integrates with Quicksight for data visualization
Used to query large S3 database
No need to spin up servers or Hadoop Clusters.

-> Query editor, saved Queries, History, Data sources, Workgroup: primary

Use Case: Analyzing CloudTrail/CloudFront/VPC/ELBlogs
Integration through ODBC/JDBC with other visualization tool
Ad-hoc logs analysis

You need to setup a crawler to run periodically, scan your data and automatically populate the Glue Data Catalogue

Cost -> $5 per TB data scanned, no charges for DDL queries and failed queries, charges for cancelled queries
Save -> Columnar formats, using partitions, compression, number of files

Athena Queries:
Query results and metadata information is stored.
You specify output folder in S3 for storing output.
Stored Procedures are not supported.
Certain DDL operation are also not supported.
Athena Federated Query (preview feature).