07. Amazon S3 Storage Classes

Amazon S3 offers multiple storage classes to optimize cost, performance, and availability for different data access patterns. Objects can be moved between these classes manually or automatically using S3 Lifecycle configurations.

---

🧩 Durability vs Availability

Concept	Description	Example
Durability	Probability that data will not be lost	99.999999999% (11 9’s) — loss of 1 object every 10,000 years (if 10M stored)
Availability	How readily accessible the service is	99.99% for S3 Standard → ~53 mins downtime/year

All S3 storage classes share the same durability, but availability varies by class.

---

💾 Amazon S3 Storage Classes

Storage Class	Availability	Description	Use Cases
S3 Standard (General Purpose)	99.99%	For frequently accessed data. Low latency, high throughput. Survives 2 concurrent AZ failures.	Big Data analytics, mobile apps, gaming, content distribution
S3 Standard-IA (Infrequent Access)	99.9%	Lower cost, for data that is rarely accessed but needs fast retrieval.	Disaster Recovery, backups
S3 One Zone-IA	99.5%	Stored in a single AZ (data lost if AZ destroyed). Lower cost.	Secondary backups, easily re-creatable data
S3 Glacier Instant Retrieval	—	Archive storage with millisecond retrieval. Min storage: 90 days.	Quarterly access backups
S3 Glacier Flexible Retrieval	—	Archive storage with multiple retrieval options: • Expedited: 1–5 mins • Standard: 3–5 hrs • Bulk: 5–12 hrs (free) Min storage: 90 days	Long-term backups, non-urgent data
S3 Glacier Deep Archive	—	Lowest-cost archive. • Standard: 12 hrs • Bulk: 48 hrs Min storage: 180 days	Compliance, regulatory archives
S3 Intelligent-Tiering	—	Automatically moves objects between access tiers based on usage. No retrieval charges. Small monthly monitoring fee.	Dynamic data sets, unknown access patterns

---

⚙️ Intelligent-Tiering Access Tiers

Tier	Type	Trigger (No Access For)	Description
Frequent Access	Automatic	Default	Standard frequent usage
Infrequent Access	Automatic	30 days	For rarely accessed data
Archive Instant Access	Automatic	90 days	Archive with millisecond retrieval
Archive Access	Optional	90–700+ days	Cold archive
Deep Archive Access	Optional	180–700+ days	Deep cold archive, lowest cost

---

🔒 Object Lock & Vault Lock

Feature	Description	Purpose
S3 Object Lock	Enforces WORM (Write Once Read Many) policy at the object level. Prevents version deletion for a specified time.	Compliance, data retention
Glacier Vault Lock	Applies WORM policy at the vault level. Locks the policy permanently (no edits allowed).	Compliance, secure archiving

---

⚙️ Amazon S3 Storage Class & Lifecycle Rules Demo

---

🖊️ S3 Classes Customize

01. Upload an Object with a Specific Storage Class

1. Open your existing S3 bucket
2. Click Upload → Add files → choose coffee.jpeg
3. Under Properties → Storage class, select one of:
   • STANDARD
   • STANDARD_IA
   • ONEZONE_IA
   • INTELLIGENT_TIERING
   • GLACIER_INSTANT_RETRIEVAL
   • GLACIER_FLEXIBLE_RETRIEVAL
   • GLACIER_DEEP_ARCHIVE
4. Click Upload

📌 Each object can have its own storage class.

02. Change Storage Class (Manual Edit)

1. Select the uploaded object
2. Go to Properties → Edit Storage Class
3. Choose a new class (e.g., from STANDARD_IA → ONEZONE_IA)
4. Click Save changes

✅ The object’s storage class is updated instantly.

---

🎰 Automate Transitions with Lifecycle Rules

1. In the same bucket, go to Management → Create Lifecycle Rule
2. Name: DemoRule
3. Scope: Apply to all objects
4. Under Transitions, define:

   After (Days)	Move To
   30	Standard-IA
   60	Intelligent-Tiering
   180	Glacier Flexible Retrieval

5. Review the settings → Save Rule

✅ S3 will automatically move objects between tiers based on their age.

---

🧠 Key Takeaways

• All S3 classes have 11 9’s durability.
• Availability decreases as cost and redundancy decrease.
• Lifecycle rules or Intelligent-Tiering can automate class transitions.
• Glacier tiers are for archival and regulatory data.
• Object Lock and Vault Lock enforce WORM compliance.
• Storage class applies per object.
• You can manually edit or automate class transitions.
• Lifecycle rules help reduce costs by archiving old data automatically.

---

⚡ Amazon S3 Express One Zone

S3 Express One Zone is a high-performance, single-AZ storage class designed for ultra-low latency and high request rates.

Objects are stored in a special Directory Bucket, not a standard S3 bucket.

⚙️ Key Characteristics

Feature	Description
Storage Type	Stored in a single Availability Zone (you choose which one)
Performance	Up to 10× faster than S3 Standard
Latency	Single-digit millisecond latency
Availability	Lower (single AZ only)
Durability	High, but confined to one AZ
Cost	~50% lower than S3 Standard
Bucket Type	Requires a Directory Bucket (not a regular bucket)

🚀 Use Cases

• Latency-sensitive workloads
• Data-intensive analytics
• AI/ML model training
• Financial modeling
• Media processing
• High-performance computing (HPC)

🔗 AWS Integrations

Best used with:

• Amazon SageMaker (training jobs)
• Amazon Athena
• Amazon EMR
• AWS Glue

---

🔐 Amazon S3 Encryption

• Server-Side Encryption (SSE):
  S3 encrypts objects after upload — default behavior.
  (Encryption done by AWS.)

• Client-Side Encryption (CSE):
  User encrypts data before upload.
  (Encryption done by user.)

✅ By default, S3 uses Server-Side Encryption for all new objects.

---

🕵️ IAM Access Analyzer for S3

• Purpose: Monitors S3 buckets for unintended public or cross-account access.
• How: Analyzes Bucket Policies, ACLs, and Access Point Policies.
• Use: Identifies buckets shared publicly or with other AWS accounts.
• Goal: Helps detect and fix security misconfigurations.