AWS Zero To Hero - CloudAdhar x TrainWithShubham
Sessions: Aug 8-9, 2026
Course sessions: Day 11-12
Exam focus: SAA-C03 Domains 1, 2, 3, and 4
Main pillars: Security, Reliability, Performance Efficiency, and Cost Optimization
Week 6 covers Amazon S3 object storage, security, lifecycle management, versioning, replication and transfer features, then compares EFS, Amazon FSx, and hybrid storage services. The labs emphasize private-by-default design, validated recovery behavior, controlled cost, and complete cleanup.
| Seq | Session | Focus | File |
|---|---|---|---|
| 01 | Day 11 | S3 foundations, storage classes, security, and cost | 01-s3-foundations-security-cost.md |
| 02 | Day 11 | Build and validate a private, versioned S3 workflow | 02-s3-security-lifecycle-lab.md |
| 03 | Day 12 | SRR, CRR, transfer, EFS, FSx, and hybrid storage | 03-s3-replication-and-hybrid-storage.md |
| 04 | Day 12 | Validate prefix replication and storage decisions | 04-s3-replication-hybrid-storage-lab.md |
| 05 | Both | Document the S3 and hybrid-storage architecture | 05-architecture-exercise.md |
| 06 | End | Remove Week 6 resources safely | 06-cleanup.md |
| 07 | End | Submit Week 6 evidence | 07-submission-format.md |
| 08 | Daily | Share learning progress | 08-linkedin-post.md |
| 09 | Review | Revise Week 6 storage decisions and exam cues | 09-quick-revision.md |
- Day 11 S3 sample files - private objects, versioning, lifecycle, storage-class, website, and Object Lock samples
- Day 12 S3 sample files - pre-rule, SRR, CRR, unmatched-prefix, transfer, multipart, hybrid-selection, and troubleshooting samples
- Explain buckets, objects, keys, prefixes, metadata, version IDs, and the Regional S3 model.
- Select S3 Standard, Intelligent-Tiering, Standard-IA, One Zone-IA, and Glacier storage classes from access, resilience, retrieval, and cost needs.
- Create private General Purpose buckets with ACLs disabled and all four Block Public Access settings enabled.
- Compare SSE-S3 and SSE-KMS and explain S3 Bucket Keys.
- Upload objects using Standard and Intelligent-Tiering.
- Create two versions of one key, observe a delete marker, and recover the object without deleting its data versions.
- Copy a private object between buckets and validate destination SSE-KMS encryption with a customer managed key.
- Prove the normal Object URL is denied and a short-lived presigned URL grants narrow temporary access.
- Configure lifecycle transitions, current and noncurrent expiration, and incomplete multipart-upload cleanup.
- Use Object Lock Legal Hold to deny deletion safely, then remove the hold and clean up the exact object version.
- Compare Same-Region Replication and Cross-Region Replication.
- Build prefix-filtered SRR from Mumbai to Mumbai and CRR from Mumbai to Tokyo.
- Observe
PENDING,COMPLETED, andREPLICAstates. - Prove that live replication is not retroactive and that later versions are replicated.
- Explain S3 Batch Replication for existing or failed eligible objects.
- Review Transfer Acceleration and its accelerated endpoint without claiming a small console upload proves a speed improvement.
- Configure lifecycle cleanup for incomplete multipart uploads.
- Review an existing EFS design and the Amazon FSx family without creating duplicate paid resources.
- Select Storage Gateway, DataSync, Snow Family, and Transfer Family from migration and hybrid-access requirements.
- Optionally complete the controlled Object Lock Compliance and native S3 HTTP website make-up labs with strict cleanup.
AWS Account - ap-south-1
Approved user -- presigned GET --> Private source S3 bucket
Anonymous user -- normal URL ----X AccessDenied
|
| CopyObject
v
Private destination bucket
SSE-KMS + S3 Bucket Key
|
v
Customer managed KMS key
Separate Object Lock bucket -- version + Legal Hold --> delete denied
Day 12
Mumbai source -- srr/ --> Mumbai SRR destination
|
`-- crr/ --> Tokyo CRR destination
Edge location -. accelerated endpoint .-> Mumbai source
Compute --> EFS or requirement-selected FSx service
On premises --> Storage Gateway / DataSync / Snow / Transfer Family
The source bucket uses SSE-S3, versioning, lifecycle rules, storage-class
examples, and Block Public Access. The destination bucket uses SSE-KMS with
alias/cloudadhar-s3-day11. Manual copy is a Day 11 operation; continuous S3
Replication belongs to Day 12.
- Source and destination bucket security settings
- Source SSE-S3 and destination SSE-KMS configuration
- S3 Bucket Key enabled on the destination
- Standard and Intelligent-Tiering object properties
- Two object versions, delete marker, and recovered object
- Successful bucket-to-bucket copy and destination encryption proof
- Normal Object URL denied
- Block Public Access safely rejecting or neutralizing the public-policy test
- Presigned access success without exposing the URL
- Enabled lifecycle rule with the complete timeline
- Object Lock Legal Hold denial and successful cleanup after hold removal
- Three-bucket Day 12 Region and security table
- Enabled
srr/andcrr/replication rules - SRR and CRR Version 1 and Version 2 evidence
- Pre-rule and unmatched-prefix objects remaining source-only
- Transfer Acceleration and incomplete multipart cleanup configuration
- EFS, FSx, and hybrid-storage decision review
- Architecture decision, cleanup proof, and public learning post
- Use a training role, not the root user.
- Upload only synthetic training data.
- Keep ACLs disabled and all four Block Public Access settings enabled.
- Never make the class bucket public to demonstrate public access.
- Treat a presigned URL as a temporary bearer credential and never publish it.
- Use Compliance mode only in the dedicated make-up bucket with a short approved retain-until date and a documented pending-cleanup plan.
- Versioning, noncurrent objects, multipart parts, storage transitions, KMS requests, and retained locked versions can create cost.
- CRR creates destination storage, requests, and inter-Region transfer cost.
- Do not enable Replication Time Control or detailed metrics for this lab.
- Do not deploy FSx, Storage Gateway, DataSync, Snow, or Transfer Family resources for a console selection review.
- A native S3 website is public HTTP. Use only an isolated disposable bucket, then remove the policy and restore every BPA control immediately.
- Empty every object version and delete marker before deleting a bucket.
- Remove S3 dependencies before disabling or scheduling deletion of the KMS key.
- Mask account IDs, ARNs, URLs, object names containing private data, access keys, session tokens, email addresses, organization data, and billing data.