João Freitas

The following article explains the new Virtual Threads feature available in Java 21 and how it can also be used in Kotlin.

It’s easy to forget, but whenever Java gains new functionality, so does Kotlin (if it uses a JVM-backend with a JDK >= 21).

JDK 21 comes with support for Virtual Threads

So Kotlin with a JDK 21 backend has support for Virtual Treads. E.g. let’s look at how we could implement a concurrentMap extension functions:

inline fun <T, R> Iterable<T>.concurrentMap(
    crossinline transform: (T) -> R
): List<R> = Executors.newVirtualThreadPerTaskExecutor().use { execService -> 
        .map { execService.submit<R> { transform(it) } }
        .map { it.get() }

The nice thing about this function is that the rest of the code doesn’t even know that Virtual Threads are used. It can certainly be optimized, but as an example it will suffice. We can test it with this code:

fun main() {
    println("Let's try out VirtualThreads!")
    val urlsToFetch: List<URI> = listOf(

    val totalTime: Duration = measureTime {
        // the urls are fetched concurrently!
        val timedResults: List<TimedValue<Result<String>>> =
            urlsToFetch.concurrentMap { fetch(it) }

        val sumOfIndividualTimesInMs = timedResults.sumOf {
        println("sum of individual execution times: ${sumOfIndividualTimesInMs}ms")
    println("total execution time: ${totalTime.inWholeMilliseconds}ms")

// operation with blocking IO
fun fetch(uri: URI): TimedValue<Result<String>> = measureTimedValue {
    runCatching {
        uri.toURL().openStream().use { it.readAllBytes().decodeToString() }

As expected the total execution time is way lower than the combined execution time its parts.

Let's try out VirtualThreads!
sum of individual execution times: 3359ms
total execution time: 1240ms

You can find the repo for this project here.

Notice that the Kotlin version used for this code is 1.9.20-Beta2. We need this high of a version since Kotlin 1.9.20 comes with a new JvmTarget for the JDK 21, so older Kotlin versions do not run on the JDK 21 backend. (At least not in Gradle.)

What about Coroutines?

But Kotlin already has Coroutines. So why would Virtual Threads be used in Kotlin anyway?

Well, both Coroutines and Virtual Threads enable concurrent programming, but according to Kotlin’s project lead, both approaches are optimized for different things and therefor either one can be the appropriate one to use depending on the usecase. Take a look at his talk at KotlinConf'23.


A rising tide (jdk version) lifts all boats (languages compiling to JVM bytecode).

#reads #stephan schröder #kotlin #java #jdk #virtual thread