Cost estimation for access patterns in DynamoDB - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ query_result

intermediate

Calculate read capacity units for strongly consistent reads

A DynamoDB table has items of size 3 KB each. You want to perform 100 strongly consistent read operations per second. How many read capacity units (RCUs) are required?

A 100 RCUs

B 300 RCUs

C 200 RCUs

D 400 RCUs

💡 Hint

Each strongly consistent read of up to 4 KB consumes 1 RCU. Reads larger than 4 KB consume additional RCUs proportionally.

Each strongly consistent read can read up to 4 KB per RCU. Since each item is 3 KB, each read consumes 1 RCU. For 100 reads per second, total RCUs = 100 * 1 = 100. But since the item size is 3 KB, which is less than 4 KB, each read consumes 1 RCU. So total is 100 RCUs.

Correction: Actually, the item size is 3 KB, so each read consumes 1 RCU. For 100 reads, 100 RCUs are needed. The correct answer is 100 RCUs.

❓ query_result

intermediate

Write capacity units for batch write operations

You want to write 50 items per second to a DynamoDB table. Each item is 2 KB in size. How many write capacity units (WCUs) do you need to provision?

A 200 WCUs

B 100 WCUs

C 25 WCUs

D 50 WCUs

💡 Hint

Each write capacity unit represents one write per second for items up to 1 KB. Larger items consume more WCUs.

Each write of up to 1 KB consumes 1 WCU. Since each item is 2 KB, each write consumes 2 WCUs. For 50 writes per second, total WCUs = 50 * 2 = 100 WCUs.

❓ optimization

advanced

Optimizing cost for eventually consistent reads

You have a table with items of size 8 KB. You need to perform 120 read operations per second. You can choose between strongly consistent or eventually consistent reads. Which option minimizes your read capacity units (RCUs) cost?

A Use eventually consistent reads, requiring 60 RCUs

B Use strongly consistent reads, requiring 240 RCUs

C Use eventually consistent reads, requiring 120 RCUs

D Use strongly consistent reads, requiring 120 RCUs

💡 Hint

Eventually consistent reads consume half the RCUs of strongly consistent reads for the same item size.

Each read of 8 KB consumes 2 RCUs for strongly consistent reads (8 KB / 4 KB = 2). For 120 reads, total RCUs = 120 * 2 = 240.

Eventually consistent reads consume half the RCUs, so 240 / 2 = 120 RCUs.

However, the option C says 60 RCUs, which is half of 120. This is incorrect. The correct calculation is 120 RCUs for eventually consistent reads.

Correction: Actually, each eventually consistent read consumes half the RCUs of strongly consistent. So for 8 KB item, strongly consistent read consumes 2 RCUs, eventually consistent consumes 1 RCU. For 120 reads, eventually consistent reads require 120 RCUs.

Therefore, option C is correct.

❓ schema

advanced

Choosing partition key to optimize cost

You have a DynamoDB table with a hot partition causing throttling. Which partition key design helps reduce cost by distributing traffic evenly?

A Use a single partition key with high traffic

B Use a partition key based on timestamp only

C Use a partition key with low cardinality

D Use a composite key with a random suffix to spread traffic

💡 Hint

Distributing traffic evenly across partitions reduces throttling and cost.

A single partition key with high traffic causes hot partitions and throttling.

Using a composite key with a random suffix spreads traffic across multiple partitions, reducing throttling and cost.

Low cardinality keys concentrate traffic, increasing throttling.

Timestamp-only keys can cause traffic spikes on recent partitions.

🔧 Debug

expert

Identify the cause of unexpectedly high write costs

A DynamoDB table shows unexpectedly high write capacity usage. The application writes 10 items per second, each 1 KB in size. The table has a global secondary index (GSI) with a write capacity of 50 WCUs. What is the likely cause of the high write costs?

A Writes to the GSI double the write capacity units consumed

B The GSI write capacity is unrelated to the main table writes

C Each write consumes only 1 WCU regardless of GSI presence

D The application is writing items larger than 1 KB

💡 Hint

Writes to a table with GSIs consume write capacity on both the table and the GSIs.

When writing to a table with GSIs, each write consumes WCUs on the main table and on each GSI.

Since the GSI has a write capacity of 50 WCUs, writes to the GSI add to the total write capacity usage.

This causes higher overall write costs than expected from the main table writes alone.

Practice

1. In DynamoDB, what primarily determines the cost of your database usage?

easy

A. The number of read and write operations per second

B. The number of tables you create

C. The size of your database backup files

D. The number of indexes you define

Solution

1. Step 1: Understand DynamoDB cost factors

   DynamoDB charges mainly based on how many read and write requests you perform per second.

2. Step 2: Identify the main cost driver

   While backups and indexes can affect cost, the primary cost is from read/write capacity usage.

3. Final Answer:

   The number of read and write operations per second -> Option A

4. Quick Check:

   Cost depends on read/write operations [OK]

Hint: Focus on read/write operations for cost estimation [OK]

Common Mistakes:

• Thinking table count directly affects cost
• Confusing backup size with operational cost
• Assuming indexes alone drive cost

2. Which of the following is the correct way to calculate the read capacity units (RCUs) for a strongly consistent read of a 3 KB item in DynamoDB?

easy

A. 4 RCUs per 3 KB read

B. 2 RCUs per 3 KB read

C. 0.5 RCU per 3 KB read

D. 1 RCU per 3 KB read

Solution

1. Step 1: Recall RCU calculation for strongly consistent reads

   One RCU allows reading up to 4 KB per second for strongly consistent reads.

2. Step 2: Calculate RCUs for 3 KB item

   Since 3 KB is less than 4 KB, it requires 1 RCU.

3. Final Answer:

   1 RCU per 3 KB read -> Option D

4. Quick Check:

   Strongly consistent read 4 KB = 1 RCU [OK]

Hint: 1 RCU per 4 KB strongly consistent read [OK]

Common Mistakes:

• Confusing strongly consistent with eventually consistent reads
• Miscalculating item size rounding
• Assuming 1 RCU per KB

3. You have a DynamoDB table with items of size 6 KB. You perform 100 eventually consistent reads per second. How many read capacity units (RCUs) do you need?

medium

A. 150 RCUs

B. 100 RCUs

C. 300 RCUs

D. 200 RCUs

Solution

1. Step 1: Calculate RCUs per read for eventually consistent reads

   One RCU allows reading 8 KB per second for eventually consistent reads, which is half the cost of strongly consistent reads.

2. Step 2: Calculate RCUs needed for 6 KB item

   6 KB item requires 2 RCUs for strongly consistent reads (since 6 KB > 4 KB, round up to 2). For eventually consistent reads, it costs half, so 1 RCU per read.

3. Step 3: Calculate total RCUs for 100 reads

   100 reads x 1 RCU = 100 RCUs.

4. Step 4: Correct calculation

   Since 6 KB is more than 4 KB, for eventually consistent reads, each read consumes ceil(6/4) = 2 RCUs * 0.5 = 1 RCU per read. So total 100 RCUs.

5. Final Answer:

   100 RCUs -> Option B

6. Quick Check:

   6 KB item eventually consistent read = 1 RCU per read [OK]

Hint: Eventually consistent reads cost half RCUs [OK]

Common Mistakes:

• Using strongly consistent read calculation for eventually consistent
• Not rounding up item size to nearest 4 KB
• Multiplying incorrectly by number of reads

4. A developer estimates write capacity units (WCUs) for 1 KB items but forgets to account for item size correctly. The code calculates 1 WCU per write regardless of size. What is the likely issue?

medium

A. Incorrectly calculating read capacity units instead

B. Overestimating WCUs for items smaller than 1 KB

C. Underestimating WCUs for items larger than 1 KB

D. No issue, 1 WCU per write is always correct

Solution

1. Step 1: Understand WCU calculation for item size

   One WCU allows writing 1 KB per second. Larger items require more WCUs, rounded up.

2. Step 2: Identify problem with fixed 1 WCU per write

   If item size > 1 KB, fixed 1 WCU underestimates needed capacity, causing throttling or errors.

3. Final Answer:

   Underestimating WCUs for items larger than 1 KB -> Option C

4. Quick Check:

   WCU depends on item size, fixed 1 WCU is wrong [OK]

Hint: Always multiply WCU by item size in KB rounded up [OK]

Common Mistakes:

• Assuming 1 WCU fits all item sizes
• Confusing read and write capacity units
• Ignoring item size when estimating capacity

5. You want to estimate the monthly cost of a DynamoDB table with 50 writes per second of 2 KB items and 100 strongly consistent reads per second of 3 KB items. The price per WCU is $0.00065 and per RCU is $0.00013. What is the approximate monthly cost (30 days)?

hard

A. $56.16

B. $18.72

C. $28.08

D. $36.00

Solution

1. Step 1: Calculate WCUs needed

   Each 2 KB write requires 2 WCUs (ceil(2/1 KB)). 50 writes/sec x 2 = 100 WCUs.

2. Step 2: Calculate RCUs needed

   Each 3 KB strongly consistent read requires 1 RCU (ceil(3/4 KB)). 100 reads/sec x 1 = 100 RCUs.

3. Step 3: Hourly cost

   WCUs: 100 x $0.00065 = $0.065. RCUs: 100 x $0.00013 = $0.013. Total: $0.078/hr.

4. Step 4: Monthly cost

   $0.078 x 24 hrs/day x 30 days = $56.16.

5. Final Answer:

   $56.16 -> Option A

6. Quick Check:

   Calculate WCUs and RCUs x price x time [OK]

Hint: Multiply capacity units by price and hours in month [OK]

Common Mistakes:

• Not rounding item size to nearest KB
• Mixing hourly and monthly rates
• Ignoring strongly consistent read cost