AWS S3 Table Buckets

Introduction

S3 Table Buckets are a new feature in Amazon S3 that allow you to store structured data as tables directly inside S3.

• Combines object storage + table format
• Uses Apache Iceberg under the hood
• Enables SQL-based querying

---

Key Concepts

Term	Description
Table Bucket	Special S3 bucket for storing tables
Namespace	Logical grouping of tables
Table	Structured dataset (rows & columns)

---

Architecture

S3 Table Bucket
└── Namespace
    ├── Table1
    └── Table2

---

Prerequisites

• AWS CLI installed and configured
• Proper IAM permissions
• Region supported for S3 Tables

---

Step 0: Create a Regular S3 Bucket

aws s3api create-bucket \
  --bucket demo-aws-s3-qt-123 \
  --region us-east-1

---

Step 1: Create a Table Bucket

aws s3tables create-table-bucket \
  --name qt-demo-s3-table \
  --region us-east-1

---

Step 2: Create Namespace and Table

List existing table buckets

aws s3tables list-table-buckets

Create a namespace

aws s3tables create-namespace \
  --table-bucket-arn arn:aws:s3tables:us-east-1:795731011370:bucket/qt-demo-s3-table \
  --namespace demo

Create a table

aws s3tables create-table \
  --table-bucket-arn arn:aws:s3tables:us-east-1:795731011370:bucket/qt-demo-s3-table \
  --namespace demo \
  --name users \
  --format ICEBERG

---

Step 3: Insert Data

Data can be inserted using:

• Apache Spark
• AWS Glue
• Data pipelines

Sample Data (JSON)

[
  { "id": 1, "name": "Alice", "age": 25 },
  { "id": 2, "name": "Bob",   "age": 30 }
]

---

Example: Insert Data Using PySpark

Install PySpark

pip install pyspark

PySpark Script

from pyspark.sql import SparkSession

spark = SparkSession.builder \
    .appName("S3TablesInsert") \
    .getOrCreate()

data = [
    (1, "Alice", 25),
    (2, "Bob",   30),
]

columns = ["id", "name", "age"]

df = spark.createDataFrame(data, columns)

# Write to S3 Table (Iceberg format)
df.writeTo("demo.users").append()

spark.stop()

---

Step 4: Query Data with Athena

SELECT * FROM users;

---

Features

• Built-in optimization (compaction, cleanup)
• Highly scalable storage
• Open table format (Apache Iceberg)
• Integration with:
  • Amazon Athena
  • Amazon Redshift
  • Apache Spark

---

Use Cases

• Data lakes
• Analytics pipelines
• Log processing
• Large-scale reporting

---

Comparison: S3 Table Buckets vs Regular S3

Feature	Regular S3	S3 Table Buckets
Storage	Files	Tables
Query	Limited	SQL supported
Format	Any	Apache Iceberg

---

Comparison: S3 Table Buckets vs DynamoDB

Feature	S3 Tables	DynamoDB
Type	Analytics	Application DB
Query	SQL	API
Scale	Very large datasets	Low-latency apps

---

Summary

• S3 Table Buckets bring table support to S3
• Ideal for analytics workloads
• Not a replacement for transactional databases like DynamoDB
• Best used with Athena or Spark

---

Useful CLI Commands

List namespaces

aws s3tables list-namespaces \
  --table-bucket-arn arn:aws:s3tables:us-east-1:795731011370:bucket/qt-demo-2205

List tables in a namespace

aws s3tables list-tables \
  --table-bucket-arn arn:aws:s3tables:us-east-1:795731011370:bucket/qt-demo-2205 \
  --namespace demo2

Delete a table

aws s3tables delete-table \
  --table-bucket-arn arn:aws:s3tables:us-east-1:795731011370:bucket/qt-demo-2205 \
  --namespace demo2 \
  --name daily_sales

Delete a namespace

aws s3tables delete-namespace \
  --table-bucket-arn arn:aws:s3tables:us-east-1:795731011370:bucket/qt-demo-2205 \
  --namespace demo2

Delete a table bucket

aws s3tables delete-table-bucket \
  --table-bucket-arn arn:aws:s3tables:us-east-1:795731011370:bucket/qt-demo-2205

Upload file to S3

aws s3 cp 02-04-2026.md s3://demo-aws-s3-qt-123/2-apr-2026.md

Put object (generic)

aws s3api put-object \
  --bucket demo-aws-s3-qt-123 \
  --key <your-key> \
  --body <your-file>