DynamoDB Transactions, Conditional Writes आणि Optimistic Locking

By Ravindra Bagale, trainer · LinkedIn · Last updated: 31 October 2026

1. रोजचे उदाहरण: दोन cashiers

दोन्ही cashiers = race. Transaction = एकत्र. Condition = stock>0. Version tag = optimistic lock.

2. Vertical steps

3. हे का लागतं?

हे का लागतं?

Concurrent writers = corruption. No idempotency = duplicates.

4. हे काय आहे?

हे काय आहे?

Conditions; Transact all-or-nothing; version attr; idempotency keys.

5. कसं करायचं

कसं करायचं (step by step)

1. Race spots.
2. Conditions.
3. Transactions.
4. Idempotency.
5. Cancel reasons.

6. Transaction

Seat update + order Put एकत्र; else काही नाही.

7. Idempotency

IDEMP# key item; not_exists create; retry Get.

8. Commands: output आणि अर्थ

aws dynamodb put-item --table-name lab-dda-shop --item '{"PK":{"S":"SEAT#A1"},"SK":{"S":"META"},"status":{"S":"FREE"},"version":{"N":"1"}}'
aws dynamodb update-item --table-name lab-dda-shop --key '{"PK":{"S":"SEAT#A1"},"SK":{"S":"META"}}' \
  --update-expression "SET #s = :taken, version = version + :one" \
  --condition-expression "#s = :free AND version = :v" \
  --expression-attribute-names '{"#s":"status"}' \
  --expression-attribute-values '{":taken":{"S":"TAKEN"},":free":{"S":"FREE"},":v":{"N":"1"},":one":{"N":"1"}}'
# Output: success once. Second identical call fails ConditionalCheckFailedException.
# Meaning: only one booker wins; version bumps.

एकच book succeed; stale version fail.

9. Industry stories

• Clinic
• Fintech
• Retail
• Travel
• Gaming

10. Lab

कसं करायचं (step by step)

1. Book+replay.
2. Transact.
3. Delete.

Cleanup

1. Delete lab.
2. Confirm.

11. Access-pattern workshop (आधी कागदावर)

1. प्रत्येक screen/API लिहा.
2. प्रत्येकासाठी key/query English मध्ये.
3. Relationships mark करा.
4. Range गरज गोल करा.
5. मगच PK/SK/GSI.
6. Peak per PK.
7. Entity JSON नमुना.
8. Peer review.

12. Copy-paste DynamoDB CLI starters

# Region habit for this series
export AWS_DEFAULT_REGION=ap-south-1

# Describe + sample
aws dynamodb describe-table --table-name ShopApp
aws dynamodb scan --table-name ShopApp --max-items 3

# Query by PK (replace names)
aws dynamodb query --table-name ShopApp   --key-condition-expression "PK = :pk"   --expression-attribute-values '{":pk":{"S":"USER#u123"}}'

# On-demand capacity mode check
aws dynamodb describe-table --table-name ShopApp   --query 'Table.BillingModeSummary'

# Streams status
aws dynamodb describe-table --table-name ShopApp   --query 'Table.StreamSpecification'

13. अजून industry stories

• Retail cart single-table
• Gaming leaderboard shard
• IoT time-series TTL
• Clinic conditional Put
• Fintech audit Streams
• Travel global tables

14. Glossary

15. Practice: मोठ्याने explain

1. 60s access patterns first.
2. 60s hot + shard.
3. 60s GSI vs LSI.
4. 60s capacity modes.
5. 60s version lock.
6. 60s Streams audit.
7. Record ×2.

16. Safe lab hygiene

• Budget alert.
• lab- prefix.
• On-demand labs.
• Delete after.
• DLQ required.
• Owner+Expiry tags.

17. या set ची links

• Intro chapter
• CRUD survey
• CLI chapter
• SDK chapter
• Serverless projects

18. Single-table sketch pad

Console आधी एक mentally table. Shop: USER#, ORD#, ITEM#, PROD#.

19. दहा oral drills

1. Access pattern?
2. PK prefixes?
3. Hot symptom?
4. Sparse GSI?
5. LSI timing?
6. On-demand bill?
7. attribute_not_exists?
8. View type trade-off?
9. Poison loop?
10. Which chapter for Put?

20. Capacity / throttle drill

1. CW metrics उघडा.
2. Mode note.
3. RCU/WCU लिहा.
4. Hottest PK.
5. Mitigation.
6. Budget alert.
7. GSI audit.
8. Streams audit.

21. Condition cheatsheet

# Create-once
attribute_not_exists(PK)

# Optimistic lock
version = :v

# Stock / seat
stock > :zero
#status = :free

# Combine
attribute_not_exists(PK) AND #type = :user

Conditions = seatbelt. SDK chapter = full samples.

22. Streams consumer checklist

• View type.
• DLQ.
• Idempotent.
• IteratorAge.
• No loop.
• Fan-out.
• Lab disable.

23. Link map batch 1

• Sequence.
• No duplicate verbs.
• Projects later.

24. अधिक oral drills

1. 45s PM.
2. Chat PK/SK.
3. 3 metrics.
4. Refuse GSI.
5. Txn story.
6. ALO.
7. Scan->Query.
8. Cleanup.

25. Teacher story: notebook → database

Ravindra Sir नी एका दुकानात सगळं एका notebook मध्ये बघितलं — customer, order, payment. unpaid bills शोधायला पाने उलटायला लागत. DynamoDB single-table म्हणजे तोच notebook नीट लेबल (PK/SK) आणि गरज असेल तरच दुसरी tab (GSI). दहा वेगळे books तयार करू नका जोपर्यंत खरंच दहा वेगळे व्यवसाय नाहीत.

Meera (clinic), Omar (travel), Aisha (retail) एकत्र design करतात तेव्हाही आधी कागद: screen यादी, key sketch, sample JSON, peer review. Console शेवटी. म्हणूनच ही deep series PutItem पुन्हा लिहित नाही — intro आणि CRUD ला link करते.

26. सोळा mini tasks (self-check)

1. Cart साठी तीन access patterns.
2. User + orders PK/SK.
3. एक hot anti-pattern.
4. 40s मध्ये write sharding.
5. GSI vs LSI without notes.
6. Sparse index projection का?
7. Flash sale mode.
8. पाच bill line items.
9. Create-once condition शब्दात.
10. Version lock वर्णन.
11. Txn कधी?
12. Audit view type.
13. दोन consumer safety rules.
14. Stream lag metric.
15. PutItem CLI कुठे?
16. Lab cleanup checklist.

27. मराठी classroom lines

• Mantra: patterns first.
• Hot = गर्दी.
• GSI vs LSI.
• Modes.
• Seatbelt.
• ALO.

28. Hindi bridge lines

• Patterns first.
• Hot load.
• GSI/LSI.
• Modes.
• Txn.
• Idempotent.

29. सामान्य errors आणि fixes

प्रत्येक row = खरी चूक. Panic करण्याआधी fix वाचा.

30. Troubleshooting checklist

1. Refresh.
2. Reasons.
3. Split.
4. Return prior.
5. Backoff.

31. FAQ

32. Interview Q&A

33. Tips — English, मराठी आणि हिंदी

उजळणी, मुलाखत आणि तुलना cards

हा lesson काही मिनिटांत revise करा. Cards phone वर save करा किंवा PNG download करा. प्रत्येक card वरचा QR हा lesson उघडतो.

अजून cards: Ravi Sir's Cards.