Docs Menu
Docs Home
/ /
Atlas Device SDKs
/ /

Threading - Swift SDK

On this page

  • Three Rules to Follow
  • Perform a Background Write
  • Wait for Async Writes to Complete
  • Commit or Cancel an Async Write
  • Communication Across Threads
  • Create a Serial Queue to use Realm on a Background Thread
  • Pass Instances Across Threads
  • Use the Same Realm Across Threads
  • Refreshing Realms
  • Frozen Objects
  • Realm's Threading Model in Depth
  • Compared and Contrasted with Git
  • Internal Structure
  • Summary
  • Sendable, Non-Sendable and Thread-Confined Types

To make your iOS and tvOS apps fast and responsive, you must balance the computing time needed to lay out the visuals and handle user interactions with the time needed to process your data and run your business logic. Typically, app developers spread this work across multiple threads: the main or UI thread for all of the user interface-related work, and one or more background threads to compute heavier workloads before sending it to the UI thread for presentation. By offloading heavy work to background threads, the UI thread can remain highly responsive regardless of the size of the workload. But it can be notoriously difficult to write thread-safe, performant, and maintainable multithreaded code that avoids issues like deadlocking and race conditions. Realm aims to simplify this for you.


See also:

As of 10.26.0, Realm provides async write methods to perform background writes. See: Perform a Background Write. With async write, you don't need to pass a thread-safe reference or frozen objects across threads.

This page describes how to manually manage realm files and objects across threads. Realm also supports using a Swift actor to manage realm access using Swift concurrency features. For an overview of Realm's actor support, refer to Use Realm with Actors - Swift SDK.

Before exploring Realm's tools for multithreaded apps, you need to understand and follow these three rules:

Don't lock to read:
Realm's Multiversion Concurrency Control (MVCC) architecture eliminates the need to lock for read operations. The values you read will never be corrupted or in a partially-modified state. You can freely read from the same Realm file on any thread without the need for locks or mutexes. Unnecessarily locking would be a performance bottleneck since each thread might need to wait its turn before reading.
Avoid synchronous writes on the UI thread if you write on a background thread:
You can write to a Realm file from any thread, but there can be only one writer at a time. Consequently, synchronous write transactions block each other. A synchronous write on the UI thread may result in your app appearing unresponsive while it waits for a write on a background thread to complete. Device Sync writes on a background thread, so you should avoid synchronous writes on the UI thread with synced realms.
Don't pass live objects, collections, or realms to other threads:
Live objects, collections, and realm instances are thread-confined: that is, they are only valid on the thread on which they were created. Practically speaking, this means you cannot pass live instances to other threads. However, Realm offers several mechanisms for sharing objects across threads.


New in version 10.26.0.

You can add, modify, or delete objects in the background using writeAsync.

With writeAsync, you don't need to pass a thread-safe reference or frozen objects across threads. Instead, call realm.writeAsync. You can provide a completion block for the method to execute on the source thread after the write completes or fails.

Things to consider when performing background writes:

  • Async writes block closing or invalidating the realm

  • You can explicitly commit or cancel transactions

let realm = try! Realm()
// Query for a specific person object on the main thread
let people = realm.objects(Person.self)
let thisPerson = people.where {
$ == "Dachary"
// Perform an async write to add dogs to that person's dog list.
// No need to pass a thread-safe reference or frozen object.
realm.writeAsync {
thisPerson?.dogs.append(objectsIn: [
Dog(value: ["name": "Ben", "age": 13]),
Dog(value: ["name": "Lita", "age": 9]),
Dog(value: ["name": "Maui", "age": 1])
} onComplete: { _ in
// Confirm the three dogs were successfully added to the person's dogs list
XCTAssertEqual(thisPerson!.dogs.count, 3)
// Query for one of the dogs we added and see that it is present
let dogs = realm.objects(Dog.self)
let benDogs = dogs.where {
$ == "Ben"
XCTAssertEqual(benDogs.count, 1)

The SDK provides a Bool to signal whether the realm is currently performing an async write. The isPerformingAsynchronousWriteOperations variable becomes true after a call to one of:

  • writeAsync

  • beginAsyncWrite

  • commitAsyncWrite

It remains true until all scheduled async write operations have completed. While this is true, this blocks closing or invalidating the realm.

To complete an async write, you or the SDK must call either:

When you use the writeAsync method, the SDK handles committing or canceling the transaction. This provides the convenience of the async write without the need to manually keep state tied to the scope of the object. However, while in the writeAsync block, you can explicitly call commitAsyncWrite or cancelAsyncWrite. If you return without calling one of these methods, writeAsync either:

  • Commits the write after executing the instructions in the write block

  • Returns an error

In either case, this completes the writeAsync operation.

For more control over when to commit or cancel the async write transaction, use the beginAsyncWrite method. When you use this method, you must explicitly commit the transactions. Returning without committing an async write cancels the transaction. beginAsyncWrite returns an ID that you can pass to cancelAsyncWrite.

commitAsyncWrite asynchronously commits a write transaction. This is the step that persists the data to the realm. commitAsyncWrite can take an onComplete block. . This block executes on the source thread once the commit completes or fails with an error.

Calling commitAsyncWrite immediately returns. This allows the caller to proceed while the SDK performs the I/O on a background thread. This method returns an ID that you can pass to cancelAsyncWrite. This cancels the pending invocation of the completion block. It does not cancel the commit itself.

You can group sequential calls to commitAsyncWrite. Batching these calls improves write performance; particularly when the batched transactions are small. To permit grouping transactions, set the isGroupingAllowed parameter to true.

You can call cancelAsyncWrite on either beginAsyncWrite or commitAsyncWrite. When you call it on beginAsyncWrite, this cancels the entire write transaction. When you call it on commitAsyncWrite, this cancels only an onComplete block you may have passed to commitAsyncWrite. It does not cancel the commit itself. You need the ID of the beginAsyncWrite or the commitAsyncWrite you want to cancel.

To access the same Realm file from different threads, you must instantiate a realm instance on every thread that needs access. As long as you specify the same configuration, all realm instances will map to the same file on disk.

One of the key rules when working with Realm in a multithreaded environment is that objects are thread-confined: you cannot access the instances of a realm, collection, or object that originated on other threads. Realm's Multiversion Concurrency Control (MVCC) architecture means that there could be many active versions of an object at any time. Thread-confinement ensures that all instances in that thread are of the same internal version.

When you need to communicate across threads, you have several options depending on your use case:

  • To modify an object on two threads, query for the object on both threads.

  • To react to changes made on any thread, use Realm's notifications.

  • To see changes that happened on another thread in the current thread's realm instance, refresh your realm instance.

  • To send a fast, read-only view of the object to other threads, "freeze" the object.

  • To keep and share many read-only views of the object in your app, copy the object from the realm.

  • To share an instance of a realm or specific object with another thread or across actor boundaries, share a thread-safe reference to the realm instance or object. For more information, refer to Pass a ThreadSafeReference.

When using Realm on a background thread, create a serial queue. Realm does not support using realms in concurrent queues, such as the global() queue.

// Initialize a serial queue, and
// perform realm operations on it
let serialQueue = DispatchQueue(label: "serial-queue")
serialQueue.async {
let realm = try! Realm(configuration: .defaultConfiguration, queue: serialQueue)
// Do something with Realm on the non-main thread

Instances of Realm, Results, List, and managed Objects are thread-confined. That means you may only use them on the thread where you created them. However, Realm provides a mechanism called thread-safe references that allows you to copy an instance created on one thread to another thread.

New in version 10.20.0: @ThreadSafe wrapper and ThreadSafeReference conform to Sendable

If you are using Swift 5.6 or higher, both the @ThreadSafe property wrapper and ThreadSafeReference conform to Sendable.

New in version 10.17.0.

You can pass thread-confined instances to another thread as follows:

  1. Use the @ThreadSafe property wrapper to declare a variable that references the original object. By definition, @ThreadSafe-wrapped variables are always optional.

  2. Pass the @ThreadSafe-wrapped variable to the other thread.

  3. Use the @ThreadSafe-wrapped variable as you would any optional. If the referenced object is removed from the realm, the referencing variable becomes nil.

let realm = try! Realm()
let person = Person(name: "Jane")
try! realm.write {
// Create thread-safe reference to person
@ThreadSafe var personRef = person
// @ThreadSafe vars are always optional. If the referenced object is deleted,
// the @ThreadSafe var will be nullified.
print("Person's name: \(personRef?.name ?? "unknown")")
// Pass the reference to a background thread
DispatchQueue(label: "background", autoreleaseFrequency: .workItem).async {
let realm = try! Realm()
try! realm.write {
// Resolve within the transaction to ensure you get the
// latest changes from other threads. If the person
// object was deleted, personRef will be nil.
guard let person = personRef else {
return // person was deleted
} = "Jane Doe"

Another way to work with an object on another thread is to query for it again on that thread. But if the object does not have a primary key, it is not trivial to query for it. You can use the @ThreadSafe wrapper on any object, regardless of whether it has a primary key.


The following example shows how to use @ThreadSafe on a function parameter. This is useful for functions that may run asynchronously or on another thread.


If your app accesses Realm in an async/await context, mark the code with @MainActor to avoid threading-related crashes.

func someLongCallToGetNewName() async -> String {
return "Janet"
func loadNameInBackground(@ThreadSafe person: Person?) async {
let newName = await someLongCallToGetNewName()
let realm = try! await Realm()
try! realm.write {
person?.name = newName
func createAndUpdatePerson() async {
let realm = try! await Realm()
let person = Person(name: "Jane")
try! realm.write {
await loadNameInBackground(person: person)
await createAndUpdatePerson()

Before Realm Swift SDK version 10.17.0 or in Objective-C, you can pass thread-confined instances to another thread as follows:

  1. Initialize a ThreadSafeReference with the thread-confined object.

  2. Pass the reference to the other thread or queue.

  3. Resolve the reference on the other thread's realm by calling Realm.resolve(_:). Use the returned object as normal.


You must resolve a ThreadSafeReference exactly once. Otherwise, the source realm remains pinned until the reference gets deallocated. For this reason, ThreadSafeReference should be short-lived.

let person = Person(name: "Jane")
let realm = try! Realm()
try! realm.write {
// Create thread-safe reference to person
let personRef = ThreadSafeReference(to: person)
// Pass the reference to a background thread
DispatchQueue(label: "background", autoreleaseFrequency: .workItem).async {
let realm = try! Realm()
try! realm.write {
// Resolve within the transaction to ensure you get the latest changes from other threads
guard let person = realm.resolve(personRef) else {
return // person was deleted
} = "Jane Doe"

Another way to work with an object on another thread is to query for it again on that thread. But if the object does not have a primary key, it is not trivial to query for it. You can use ThreadSafeReference on any object, regardless of whether it has a primary key. You can also use it with lists and results.

The downside is that ThreadSafeReference requires some boilerplate. You must remember to wrap everything in a DispatchQueue with a properly-scoped autoreleaseFrequency so the objects do not linger on the background thread. So, it can be helpful to make a convenience extension to handle the boilerplate as follows:

extension Realm {
func writeAsync<T: ThreadConfined>(_ passedObject: T, errorHandler: @escaping ((_ error: Swift.Error) -> Void) = { _ in return }, block: @escaping ((Realm, T?) -> Void)) {
let objectReference = ThreadSafeReference(to: passedObject)
let configuration = self.configuration
DispatchQueue(label: "background", autoreleaseFrequency: .workItem).async {
do {
let realm = try Realm(configuration: configuration)
try realm.write {
// Resolve within the transaction to ensure you get the latest changes from other threads
let object = realm.resolve(objectReference)
block(realm, object)
} catch {

This extension adds a writeAsync() method to the Realm class. This method passes an instance to a background thread for you.


Suppose you made an email app and want to delete all read emails in the background. You can now do it with two lines of code. Note that the closure runs on the background thread and receives its own version of both the realm and passed object:

let realm = try! Realm()
let readEmails = realm.objects(Email.self).where {
$ == true
realm.writeAsync(readEmails) { (realm, readEmails) in
guard let readEmails = readEmails else {
// Already deleted

You cannot share realm instances across threads.

To use the same Realm file across threads, open a different realm instance on each thread. As long as you use the same configuration, all Realm instances will map to the same file on disk.

When you open a realm, it reflects the most recent successful write commit and remains on that version until it is refreshed. This means that the realm will not see changes that happened on another thread until the next refresh. A realm on the UI thread -- more precisely, on any event loop thread -- automatically refreshes itself at the beginning of that thread's loop. However, you must manually refresh realm instances that do not exist on loop threads or that have auto-refresh disabled.

if (![realm autorefresh]) {
[realm refresh]
if (!realm.autorefresh) {
// Manually refresh

Live, thread-confined objects work fine in most cases. However, some apps -- those based on reactive, event stream-based architectures, for example -- need to send immutable copies around to many threads for processing before ultimately ending up on the UI thread. Making a deep copy every time would be expensive, and Realm does not allow live instances to be shared across threads. In this case, you can freeze and thaw objects, collections, and realms.

Freezing creates an immutable view of a specific object, collection, or realm. The frozen object, collection, or realm still exists on disk, and does not need to be deeply copied when passed around to other threads. You can freely share the frozen object across threads without concern for thread issues. When you freeze a realm, its child objects also become frozen.


Use ThreadSafeReference with Swift Actors

Realm does not currently support using thaw() with Swift Actors. To work with Realm data across actor boundaries, use ThreadSafeReference instead of frozen objects. For more information, refer to Pass a ThreadSafeReference.

Frozen objects are not live and do not automatically update. They are effectively snapshots of the object state at the time of freezing. Thawing an object returns a live version of the frozen object.

// Get an immutable copy of the realm that can be passed across threads
RLMRealm *frozenRealm = [realm freeze];
RLMResults *dogs = [Dog allObjectsInRealm:realm];
// You can freeze collections
RLMResults *frozenDogs = [dogs freeze];
// You can still read from frozen realms
RLMResults *frozenDogs2 = [Dog allObjectsInRealm:frozenRealm];
Dog *dog = [dogs firstObject];
// You can freeze objects
Dog *frozenDog = [dog freeze];
// To modify frozen objects, you can thaw them
// You can thaw collections
RLMResults *thawedDogs = [dogs thaw];
// You can thaw objects
Dog *thawedDog = [dog thaw];
// You can thaw frozen realms
RLMRealm *thawedRealm = [realm thaw];
let realm = try! Realm()
// Get an immutable copy of the realm that can be passed across threads
let frozenRealm = realm.freeze()
let people = realm.objects(Person.self)
// You can freeze collections
let frozenPeople = people.freeze()
// You can still read from frozen realms
let frozenPeople2 = frozenRealm.objects(Person.self)
let person = people.first!
// You can freeze objects
let frozenPerson = person.freeze()
// Frozen objects have a reference to a frozen realm

When working with frozen objects, an attempt to do any of the following throws an exception:

  • Opening a write transaction on a frozen realm.

  • Modifying a frozen object.

  • Adding a change listener to a frozen realm, collection, or object.

You can use isFrozen to check if the object is frozen. This is always thread-safe.

if ([realm isFrozen]) {
// ...
if (realm.isFrozen) {
// ...

Frozen objects remain valid as long as the live realm that spawned them stays open. Therefore, avoid closing the live realm until all threads are done with the frozen objects. You can close a frozen realm before the live realm is closed.


On caching frozen objects

Caching too many frozen objects can have a negative impact on the realm file size. "Too many" depends on your specific target device and the size of your Realm objects. If you need to cache a large number of versions, consider copying what you need out of the realm instead.

To modify a frozen object, you must thaw the object. Alternately, you can query for it on an unfrozen realm, then modify it. Calling thaw on a live object, collection, or realm returns itself.

Thawing an object or collection also thaws the realm it references.

// Read from a frozen realm
let frozenPeople = frozenRealm.objects(Person.self)
// The collection that we pull from the frozen realm is also frozen
// Get an individual person from the collection
let frozenPerson = frozenPeople.first!
// To modify the person, you must first thaw it
// You can also thaw collections and realms
let thawedPerson = frozenPerson.thaw()
// Check to make sure this person is valid. An object is
// invalidated when it is deleted from its managing realm,
// or when its managing realm has invalidate() called on it.
assert(thawedPerson?.isInvalidated == false)
// Thawing the person also thaws the frozen realm it references
assert(thawedPerson!.realm!.isFrozen == false)
// Let's make the code easier to follow by naming the thawed realm
let thawedRealm = thawedPerson!.realm!
// Now, you can modify the todo
try! thawedRealm.write {
thawedPerson!.name = "John Michael Kane"

When you append to a frozen collection, you must thaw both the collection and the object that you want to append. In this example, we query for two objects in a frozen Realm:

  • A Person object that has a List property of Dog objects

  • A Dog object

We must thaw both objects before we can append the Dog to the Dog List collection on the Person. If we thaw only the Person object but not the Dog, Realm throws an error.

The same rule applies when passing frozen objects across threads. A common case might be calling a function on a background thread to do some work instead of blocking the UI.

// Get a copy of frozen objects.
// Here, we're getting them from a frozen realm,
// but you might also be passing them across threads.
let frozenTimmy = frozenRealm.objects(Person.self).where {
$ == "Timmy"
let frozenLassie = frozenRealm.objects(Dog.self).where {
$ == "Lassie"
// Confirm the objects are frozen.
assert(frozenTimmy.isFrozen == true)
assert(frozenLassie.isFrozen == true)
// Thaw the frozen objects. You must thaw both the object
// you want to append and the collection you want to append it to.
let thawedTimmy = frozenTimmy.thaw()
let thawedLassie = frozenLassie.thaw()
let realm = try! Realm()
try! realm.write {
XCTAssertEqual(thawedTimmy?.dogs.first?.name, "Lassie")

Realm provides safe, fast, lock-free, and concurrent access across threads with its Multiversion Concurrency Control (MVCC) architecture.

If you are familiar with a distributed version control system like Git, you may already have an intuitive understanding of MVCC. Two fundamental elements of Git are:

  • Commits, which are atomic writes.

  • Branches, which are different versions of the commit history.

Similarly, Realm has atomically-committed writes in the form of transactions. Realm also has many different versions of the history at any given time, like branches.

Unlike Git, which actively supports distribution and divergence through forking, a realm only has one true latest version at any given time and always writes to the head of that latest version. Realm cannot write to a previous version. This means your data converges on one latest version of the truth.

A realm is implemented using a B+ tree data structure. The top-level node represents a version of the realm; child nodes are objects in that version of the realm. The realm has a pointer to its latest version, much like how Git has a pointer to its HEAD commit.

Realm uses a copy-on-write technique to ensure isolation and durability. When you make changes, Realm copies the relevant part of the tree for writing. Realm then commits the changes in two phases:

  • Realm writes changes to disk and verifies success.

  • Realm then sets its latest version pointer to point to the newly-written version.

This two-step commit process guarantees that even if the write failed partway, the original version is not corrupted in any way because the changes were made to a copy of the relevant part of the tree. Likewise, the realm's root pointer will point to the original version until the new version is guaranteed to be valid.


The following diagram illustrates the commit process:

Realm copies the relevant part of the tree for writes, then replaces the latest version by updating a pointer.
click to enlarge
  1. The realm is structured as a tree. The realm has a pointer to its latest version, V1.

  2. When writing, Realm creates a new version V2 based on V1. Realm makes copies of objects for modification (A 1, C 1), while links to unmodified objects continue to point to the original versions (B, D).

  3. After validating the commit, Realm updates the pointer to the new latest version, V2. Realm then discards old nodes no longer connected to the tree.

Realm uses zero-copy techniques like memory mapping to handle data. When you read a value from the realm, you are virtually looking at the value on the actual disk, not a copy of it. This is the basis for live objects. This is also why a realm head pointer can be set to point to the new version after the write to disk has been validated.

  • Realm enables simple and safe multithreaded code when you follow three rules:

    • don't lock to read

    • avoid writes on the UI thread if you write on background threads or use Device Sync

    • don't pass live objects to other threads.

  • There is a proper way to share objects across threads for each use case.

  • In order to see changes made on other threads in your realm instance, you must manually refresh realm instances that do not exist on "loop" threads or that have auto-refresh disabled.

  • For apps based on reactive, event-stream-based architectures, you can freeze objects, collections, and realms in order to pass shallow copies around efficiently to different threads for processing.

  • Realm's multiversion concurrency control (MVCC) architecture is similar to Git's. Unlike Git, Realm has only one true latest version for each realm.

  • Realm commits in two stages to guarantee isolation and durability.

The Realm Swift SDK public API contains types that fall into three broad categories:

  • Sendable

  • Not Sendable and not thread confined

  • Thread-confined

You can share types that are not Sendable and not thread confined between threads, but you must synchronize them.

Thread-confined types, unless frozen, are confined to an isolation context. You cannot pass them between these contexts even with synchronization.





Filter Data