Cloud Functions pricing in GCP - Step-by-Step Execution

Process Flow - Cloud Functions pricing

Function Triggered

↓

Count Invocations

↓

Measure Execution Time

↓

Calculate Memory Used

↓

Apply Free Tier Credits

↓

Calculate Cost

↓

Bill User

When a cloud function runs, GCP counts how many times it runs, how long it runs, and how much memory it uses. Then it subtracts free usage and calculates the cost.

Execution Sample

GCP

Trigger function
Count invocation = 1
Execution time = 200ms
Memory used = 256MB
Calculate cost
Apply free tier
Bill user

This traces a single function execution and how pricing is calculated step-by-step.

Process Table

Step	Action	Value	Calculation	Result
1	Function triggered	Invocation count +1	Invocation count = 1	Invocation count = 1
2	Measure execution time	200 ms	Execution time = 200 ms	Execution time = 200 ms
3	Measure memory used	256 MB	Memory used = 256 MB	Memory used = 256 MB
4	Calculate GB-seconds	0.2 sec * 0.256 GB	0.0512 GB-seconds	0.0512 GB-seconds
5	Apply free tier	Free 2 million invocations, 400,000 GB-seconds	No cost for this invocation	Cost = $0
6	Calculate cost	Invocation cost + compute cost	0 invocations charged, 0 compute charged	Total cost = $0
7	Bill user	Total cost	Charge $0 for this invocation	User billed $0

💡 Free tier covers this invocation, so no cost is charged.

Status Tracker

Variable	Start	After Step 1	After Step 2	After Step 3	After Step 4	After Step 5	Final
Invocation count	0	1	1	1	1	1	1
Execution time (sec)	0	0	0.2	0.2	0.2	0.2	0.2
Memory used (GB)	0	0	0	0.256	0.256	0.256	0.256
GB-seconds	0	0	0	0	0.0512	0.0512	0.0512
Cost ($)	0	0	0	0	0	0	0

Key Moments - 2 Insights

Why is the cost zero even though the function ran?

Because the free tier covers 2 million invocations and 400,000 GB-seconds, this single invocation is free as shown in step 5 of the execution table.

How is GB-seconds calculated?

GB-seconds is execution time in seconds multiplied by memory used in GB, as shown in step 4 where 0.2 seconds * 0.256 GB = 0.0512 GB-seconds.

Visual Quiz - 3 Questions

Look at the execution table, what is the invocation count after step 1?

A 2

B 0

C 1

D 256

💡 Hint

Check the 'Invocation count' variable in the execution_table row for step 1.

At which step is the free tier applied to reduce cost?

A Step 3

B Step 5

C Step 2

D Step 7

💡 Hint

Look for the step mentioning 'Apply free tier' in the execution_table.

If the execution time doubled, how would the GB-seconds value change at step 4?

A It would double

B It would halve

C It would stay the same

D It would be zero

💡 Hint

GB-seconds = execution time * memory used, so doubling time doubles GB-seconds (see variable_tracker).

Concept Snapshot

Cloud Functions pricing counts how many times your function runs (invocations), how long it runs (execution time), and how much memory it uses.
Pricing is based on GB-seconds (memory * time) and number of invocations.
Google Cloud offers a free tier: 2 million invocations and 400,000 GB-seconds free each month.
Costs apply only after free tier is exceeded.
Billing happens after calculating usage minus free tier credits.

Full Transcript

When you run a Cloud Function, Google Cloud counts how many times it runs, how long it runs, and how much memory it uses. It multiplies execution time by memory to get GB-seconds. Then it subtracts the free tier amounts. If your usage is within free limits, you pay nothing. Otherwise, you pay for extra invocations and GB-seconds. This example shows a function running once for 200 milliseconds using 256 MB memory. The free tier covers this usage, so the cost is zero. The key steps are counting invocations, measuring time and memory, calculating GB-seconds, applying free tier, and billing the user.

Practice

1. What are the main factors that determine the cost of Google Cloud Functions?

easy

A. Number of API calls, database queries, and cache hits

B. Number of invocations, compute time, and outbound data

C. CPU type, memory size, and disk I/O

D. Number of users, storage size, and network speed

Solution

1. Step 1: Identify pricing components

   Google Cloud Functions pricing is based on how many times functions run (invocations), how long they run (compute time), and how much data they send out (outbound data).

2. Step 2: Match factors to options

   Number of invocations, compute time, and outbound data lists these exact factors, while others mention unrelated metrics like users or storage.

3. Final Answer:

   Number of invocations, compute time, and outbound data -> Option B

4. Quick Check:

   Pricing factors = invocations, compute time, outbound data [OK]

Hint: Remember: cost depends on usage, not users or storage [OK]

Common Mistakes:

• Confusing storage or users with function pricing
• Thinking CPU type alone affects cost
• Mixing up inbound and outbound data charges

2. Which of the following is the correct way to describe the free tier for Google Cloud Functions?

easy

A. It offers unlimited invocations and compute time for all users

B. It charges a small fee for every invocation regardless of usage

C. It provides a fixed number of free invocations and compute time each month

D. It only applies to functions running in specific regions

Solution

1. Step 1: Understand free tier basics

   The free tier gives a set amount of free invocations and compute time monthly, not unlimited usage.

2. Step 2: Compare options

   It provides a fixed number of free invocations and compute time each month correctly states the free tier provides fixed free usage. It offers unlimited invocations and compute time for all users is wrong because it says unlimited. It charges a small fee for every invocation regardless of usage is wrong because free tier means no charge up to limits. It only applies to functions running in specific regions is incorrect as free tier applies globally.

3. Final Answer:

   It provides a fixed number of free invocations and compute time each month -> Option C

4. Quick Check:

   Free tier = fixed free usage monthly [OK]

Hint: Free tier means limited free usage, not unlimited [OK]

Common Mistakes:

• Assuming free tier is unlimited
• Thinking free tier applies only in some regions
• Believing every invocation is charged

3. A Cloud Function runs 1,000,000 times in a month, each execution lasting 200 milliseconds with 256 MB memory allocated. Which factor primarily affects the compute time cost?

medium

A. Number of invocations only

B. Outbound data volume

C. Function trigger type

D. Execution duration and memory size

Solution

1. Step 1: Identify compute time cost factors

   Compute time cost depends on how long the function runs and how much memory it uses during execution.

2. Step 2: Analyze options

   Execution duration and memory size correctly identifies execution duration and memory size as key for compute cost. Invocations count affects invocation cost, not compute time alone. Outbound data and trigger type do not affect compute time cost.

3. Final Answer:

   Execution duration and memory size -> Option D

4. Quick Check:

   Compute cost = duration x memory [OK]

Hint: Compute cost = run time x memory size [OK]

Common Mistakes:

• Confusing invocation count with compute time cost
• Ignoring memory size in cost calculation
• Thinking trigger type affects compute cost

4. You notice your Cloud Function costs are unexpectedly high. Which of the following is the most likely cause?

medium

A. Function runs longer than expected or uses more memory

B. You exceeded the free tier for outbound data only

C. You deployed the function in multiple regions accidentally

D. You used a deprecated runtime version

Solution

1. Step 1: Identify common cost drivers

   High costs usually come from functions running longer or using more memory, increasing compute time charges.

2. Step 2: Evaluate other options

   You exceeded the free tier for outbound data only is less likely since outbound data is often smaller cost. You deployed the function in multiple regions accidentally might increase cost but is less common. You used a deprecated runtime version affects functionality, not pricing directly.

3. Final Answer:

   Function runs longer than expected or uses more memory -> Option A

4. Quick Check:

   Longer run or more memory = higher cost [OK]

Hint: Check function duration and memory to find cost spikes [OK]

Common Mistakes:

• Blaming deprecated runtime for cost increase
• Ignoring function execution time impact
• Assuming outbound data is main cost driver

5. You want to optimize costs for a Cloud Function that processes large files and sends results over the network. Which combination of strategies will most effectively reduce your bill?

hard

A. Reduce function memory allocation and minimize outbound data size

B. Increase function timeout and add more retries

C. Deploy multiple identical functions in different regions

D. Use a higher memory allocation to speed up execution regardless of data size

Solution

1. Step 1: Identify cost factors for large file processing

   Costs come from compute time (affected by memory and duration) and outbound data size.

2. Step 2: Evaluate optimization strategies

   Reducing memory allocation lowers compute cost if performance is sufficient. Minimizing outbound data reduces network charges. Increasing timeout or retries raises cost. Deploying multiple functions increases cost. Higher memory may speed execution but can increase cost if not balanced.

3. Final Answer:

   Reduce function memory allocation and minimize outbound data size -> Option A

4. Quick Check:

   Lower memory + less data = cost savings [OK]

Hint: Cut memory and data size to save costs effectively [OK]

Common Mistakes:

• Thinking more memory always reduces cost
• Adding retries without cost consideration
• Deploying duplicates increases cost unnecessarily