Structured concurrency model in Swift - Step-by-Step Execution

Concept Flow - Structured concurrency model

Start Task

↓

Create Child Tasks

↓

Run Child Tasks Concurrently

↓

Wait for All Child Tasks

↓

Collect Results

↓

End Task

Structured concurrency runs child tasks inside a parent task, waits for all to finish, then continues.

Execution Sample

Swift

func fetchData() async -> String {
  async let a = fetchA()
  async let b = fetchB()
  return await a + b
}

Runs fetchA and fetchB concurrently, waits for both, then combines results.

Execution Table

Step	Action	Task State	Result/Output
1	Start fetchData()	Parent task started	No output yet
2	Create async let a = fetchA()	Child task A started	No output yet
3	Create async let b = fetchB()	Child task B started	No output yet
4	Run fetchA() concurrently	Child task A running	No output yet
5	Run fetchB() concurrently	Child task B running	No output yet
6	Await a and b results	Waiting for child tasks	No output yet
7	fetchA() completes	Child task A done	Result A ready
8	fetchB() completes	Child task B done	Result B ready
9	Combine results a + b	Parent task resumes	Combined result returned
10	fetchData() ends	Parent task done	Returns combined string

💡 All child tasks complete, parent task returns combined result.

Variable Tracker

Variable	Start	After Step 7	After Step 8	Final
a	Not started	Result A ready	Result A ready	Result A ready
b	Not started	Not ready	Result B ready	Result B ready
Combined Result	None	None	None	Result A + Result B

Key Moments - 3 Insights

Why do child tasks run concurrently instead of one after another?

Because async let starts child tasks immediately and runs them in parallel, as shown in steps 4 and 5.

When does the parent task wait for child tasks to finish?

At step 6, the parent uses await to pause until both child tasks complete, ensuring safe result collection.

What happens if one child task fails?

Structured concurrency cancels sibling tasks and propagates the error to the parent, maintaining task safety.

Visual Quiz - 3 Questions

Look at the execution table, at which step does the parent task wait for child tasks to finish?

A Step 8

B Step 3

C Step 6

D Step 10

💡 Hint

Check the 'Action' column for when 'Await a and b results' happens.

According to variable_tracker, what is the value of variable 'b' after step 7?

A Result B ready

B Not ready

C Result A ready

D None

💡 Hint

Look at the 'After Step 7' column for variable 'b' in variable_tracker.

If fetchB() took longer, which step would be delayed in the execution_table?

A Step 8

B Step 7

C Step 9

D Step 4

💡 Hint

Step 8 shows fetchB() completion; delay affects this step.

Concept Snapshot

Structured concurrency in Swift:
- Use async let to start child tasks concurrently.
- Parent task waits with await for all children.
- Ensures safe, predictable task lifetimes.
- Errors in children cancel siblings and propagate.
- Simplifies async code by structuring task hierarchy.

Full Transcript

Structured concurrency in Swift means running child tasks inside a parent task. The parent starts child tasks using async let, which runs them at the same time. Then the parent waits for all child tasks to finish using await. This way, the parent collects results safely and continues. If any child task fails, the others are canceled and the error goes to the parent. This model keeps async code organized and predictable.

Practice

1. What is the main purpose of structured concurrency in Swift?

easy

A. To group tasks so they run safely and complete before continuing

B. To allow tasks to run forever without waiting

C. To make tasks run only on the main thread

D. To avoid using async and await keywords

Solution

1. Step 1: Understand structured concurrency concept

   Structured concurrency groups related tasks so they run safely and finish before the program moves on.

2. Step 2: Compare options with concept

   Only To group tasks so they run safely and complete before continuing correctly describes grouping tasks to run safely and complete before continuing.

3. Final Answer:

   To group tasks so they run safely and complete before continuing -> Option A

4. Quick Check:

   Structured concurrency = safe grouped tasks [OK]

Hint: Structured concurrency means tasks finish before moving on [OK]

Common Mistakes:

• Thinking tasks run forever without waiting
• Believing tasks only run on main thread
• Confusing structured concurrency with avoiding async/await

2. Which of the following is the correct syntax to create a task group in Swift?

easy

A. withTaskGroup(of: Int.self) { group in ... }

B. asyncGroup { ... }

C. taskGroup.run { ... }

D. TaskGroup { await ... }

Solution

1. Step 1: Recall correct task group syntax

   Swift uses withTaskGroup(of: Type.self) { group in ... } to create a task group.

2. Step 2: Check options for exact syntax

   Only withTaskGroup(of: Int.self) { group in ... } matches the correct syntax; others are invalid or made-up.

3. Final Answer:

   withTaskGroup(of: Int.self) { group in ... } -> Option A

4. Quick Check:

   Task group syntax = withTaskGroup(of:) [OK]

Hint: Task groups start with withTaskGroup(of:) [OK]

Common Mistakes:

• Using TaskGroup without 'with'
• Trying taskGroup.run which doesn't exist
• Using asyncGroup which is not a Swift keyword

3. What will be the output of this Swift code using a task group?

func fetchValues() async -> [Int] {
  await withTaskGroup(of: Int.self) { group in
    group.addTask { 3 }
    group.addTask { 7 }
    group.addTask { 2 }

    var results = [Int]()
    for await value in group {
      results.append(value)
    }
    return results.sorted()
  }
}

Task {
  let values = await fetchValues()
  print(values)
}

medium

A. [3, 7, 2]

B. [2, 3, 7]

C. [7, 3, 2]

D. Compilation error

Solution

1. Step 1: Understand task group execution

   The tasks return 3, 7, and 2 asynchronously. The for-await loop collects them in any order.

2. Step 2: Analyze the return statement

   The results array is sorted before returning, so output is sorted ascending: [2, 3, 7].

3. Final Answer:

   [2, 3, 7] -> Option B

4. Quick Check:

   Sorted results = [2, 3, 7] [OK]

Hint: Remember results are sorted before printing [OK]

Common Mistakes:

• Assuming tasks run in order added
• Forgetting the sorting step
• Thinking code causes compile error

4. Identify the error in this Swift code using a task group:

func calculateSum() async -> Int {
  await withTaskGroup(of: Int.self) { group in
    group.addTask { 10 }
    group.addTask { 20 }
    var total = 0
    for value in group {
      total += value
    }
    return total
  }
}

medium

A. Return type should be Void

B. Cannot add tasks inside withTaskGroup

C. Missing 'await' before 'for' loop

D. Variable 'total' should be declared outside the group

Solution

1. Step 1: Check for correct async iteration

   When iterating over a task group, 'for await' must be used to await each task's result.

2. Step 2: Identify missing 'await'

   The code uses 'for value in group' instead of 'for await value in group', causing a syntax error.

3. Final Answer:

   Missing 'await' before 'for' loop -> Option C

4. Quick Check:

   Async iteration requires 'for await' [OK]

Hint: Use 'for await' to loop over task groups [OK]

Common Mistakes:

• Using 'for' instead of 'for await' on task groups
• Thinking tasks can't be added inside withTaskGroup
• Misunderstanding return type requirements

5. You want to fetch user data and user posts concurrently using structured concurrency. Which approach correctly uses a task group to run both tasks and combine their results?

hard

A. Use DispatchQueue.global().async to run tasks and wait with sleep()

B. Run tasks sequentially without a task group to avoid concurrency issues

C. Create two separate Task blocks and return results without waiting

D. Use withTaskGroup to add two tasks fetching data and posts, then await all and return combined tuple

Solution

1. Step 1: Understand concurrency for multiple async tasks

   Structured concurrency uses task groups to run multiple async tasks concurrently and wait for all to finish.

2. Step 2: Evaluate options for correct concurrency usage

   Use withTaskGroup to add two tasks fetching data and posts, then await all and return combined tuple correctly uses withTaskGroup to add tasks, await their completion, and combine results safely.

3. Final Answer:

   Use withTaskGroup to add two tasks fetching data and posts, then await all and return combined tuple -> Option D

4. Quick Check:

   Task groups run and wait for multiple tasks [OK]

Hint: Use withTaskGroup to run and wait multiple tasks [OK]

Common Mistakes:

• Running tasks sequentially losing concurrency benefits
• Using DispatchQueue with sleep instead of structured concurrency
• Not waiting for tasks to finish before returning