The first language for Microservices

Built for the networked age: distributable by design

Jolie code is always contained in services, which you can always move from being local to remote and vice versa, without altering the logic of your programs. A monolithical application can scale to being distributed by design: if you decide to distribute a part of it, just take it and execute it in another machine.

Program your microservice system

Deploy it in a single machine

Deploy it in two different machines

Deploy it in four different machines

Everything you build can be used to build again

Jolie offers many ways for building complex software from simple services. Even the deployment architecture of a system can be programmed with native primitives, generalising common practices. Whatever you build, is again a service that you can expose; so, it can be reused to build again! Here are some examples of composition:

Orchestration: an orchestrator is a service that offers functionalities obtained by coordinating other services with a workflow.

Aggregation: a generalisation of proxies and load balancers, which you can use to compose and expose the APIs of separate services.

Redirection: a generalisation of virtual servers, which hides the actual locations of services to clients by assigning logical names to services.

Embedding: a generalisation of application servers, which runs other services as inner components. It enables fast local communications and can even run code written in different languages than Jolie, such as Java and Javascript (with more coming)!

Structured workflows

Jolie comes with native primitives for structuring workflows, for example in sequences (one after the other) or parallels (go at the same time). This makes the code follow naturally from the requirements, avoiding error-prone bookkeeping variables for checking what happened so far in a computation. For example, the following code says that the operations publish and edit become available at the same time (|), but only after (;) operation login is invoked:

login( credentials )() { checkCredentials };
{ publish( x ) | edit( y ) }

Dynamic error handling for parallel code

Programming reliable parallel code is challenging because faults may cause side-effects in parallel activities. Jolie comes with a solid semantics for parallel fault handling. Programmers can update the behaviour of fault handlers at runtime, following the execution of activities thanks to the install primitive.

include "console.iol"
include "time.iol"

main
{
    scope( grandFather )
    {
        install( this => println@Console( "recovering grandFather" )() );
        scope( father )
        {
            install( this => println@Console( "recovering father" )() );
            scope ( son )
            {
                install( this => println@Console( "recovering son" )() );
                sleep@Time( 500 )();
                println@Console( "Son's code block" )()
            }
        }
    }
    |
    throw( a_fault )
}