Repository files navigation

Myriad - microservice coördination

use Myriad;Myriad->new->run;

Myriad provides a framework for dealing with asynchronous, microservice-based code.It is intended for use in an environment such as Kubernetes to support horizontalscaling for larger systems.

Overall this framework encourages - but does not enforce - single-responsibilityin each microservice: each service should integrate with at most one external system,and integration should be kept in separate services from business logic or aggregation.This is at odds with common microservice frameworks, so perhaps it would be more accurateto say that this framework is aimed at developing "nanoservices" instead.

If you expect to be dealing with more traffic than a single server can handle,or you have a development team larger than 30-50 or so, this might be of interest.

For a smaller system with a handful of users, it'sprobably overkill!

• Myriad::Service - load this in your own code to turn it into a microservice
• Myriad::RPC - the RPC abstraction layer, in$self->rpc
• Myriad::Storage - abstraction layer for storage, available as$self->storage within services
• Myriad::Subscription - the subscription handling layer, in$self->subscription

Each of the three abstractions has various implementations. You'd set one on startupand that would provide functionality through the top-level abstraction layer. Service codegenerally shouldn't need to care which implementation is applied. There may however be caseswhere transactional behaviour differs between implementations, so there is some basicfunctionality planned for checking whether RPC/storage/subscription use the same underlyingmechanism for transactional safety.

TheMyriad::Storage abstract API is a good starting point here.

For storage implementations, we have:

• Myriad::Storage::Redis
• Myriad::Storage::PostgreSQL
• Myriad::Storage::Memory

Additional transport mechanisms may be available, see CPAN for details.

Simple request/response patterns are handled with theMyriad::RPC layer ("remote procedure call").

Details on the request are inMyriad::RPC::Request and the response to be sent back is inMyriad::RPC::Response.

• Myriad::RPC::Implementation::Redis
• Myriad::RPC::Implementation::PostgreSQL
• Myriad::RPC::Implementation::Memory

Additional transport mechanisms may be available, see CPAN for details.

TheMyriad::Subscription abstraction layer defines the available API here.

Subscription implementations include:

• Myriad::Subscription::Implementation::Redis
• Myriad::Subscription::Implementation::PostgreSQL
• Myriad::Subscription::Implementation::Memory

Additional transport mechanisms may be available, see CPAN for details.

Note thatsome layers don't have implementations for all transports - MQ for example does not really provide a concept of "storage".

Each of these implementations is supposed to separate out the logic from the actual transport calls, so there's a separate ::Transport set of classes here:

• Myriad::Transport::Redis
• Myriad::Transport::PostgreSQL
• Myriad::Transport::Memory

which deal with the lower-level interaction with the protocol, connection management and so on. More details on thatcan be found inMyriad::Transport - but it's typically only useful for people working on theMyriad implementation itself.

Documentation for these classes may also be of use:

• Myriad::Exception - generic errors, provides"throw" in Myriad::Exception and we recommend that all service errors implement this rôle
• Myriad::Plugin - adds specific functionality to services
• Myriad::Bootstrap - startup used inmyriad.pl for providing autorestart and other functionality
• Myriad::Service - base class for a service
• Myriad::Registry - support for registering services and methods within the current process
• Myriad::Config - general config support, commandline/file/storage

Returns the mainIO::Async::Loop instance for this process.

Hashref of services that have been added to this instance,asname =>Myriad::Service pairs.

Applies configuration from commandline parameters.

Expects a list of parameters and applies the following logic for each one:

• if it contains:: and a wildcard*, it's treated as a service module base name, and allmodules under that immediate namespace will be loaded
• if it contains::, it's treated as a comma-separated list of service module names to load
• a- prefix is a standard getopt parameter

Returns theMyriad::Transport instance according to the config value.

it's designed to be used by tests, so be careful before using it in the framework code.

it takes a single param

• component - the RPC, Subscription or storage in lower case

TheNet::Async::Redis (or compatible) instance used for service coördination.

TheMyriad::Transport::Memory instance.

TheMyriad::RPC instance to serve RPC requests.

TheMyriad::RPC::Client instance to request other services RPC.

TheNet::Async::HTTP::Server (or compatible) instance used for health checksand metrics.

TheMyriad::Subscription instance to manage events.

TheMyriad::Storage instance to manage data.

Returns the commonMyriad::Registry representing the current service state.

Instantiates and adds a new service to the"loop".

Returns the service instance.

Looks up the given service, returning the instance if it exists.

Will throw an exception if the service cannot be found.

a source to corresponde to any high level events.

Requests shutdown.

Registers a coderef to be called during startup.The coderef is expected to return aFuture.

Registers a coderef to be called during shutdown.

The coderef is expected to return aFuture indicating completion.

Returns a copy of the runFuture.

This would resolve once the process is running and it'sready to accept requests.

Returns a copy of the shutdownFuture.

This would resolve once the process is about to shut down,triggered by a fault or a Unix signal.

Prepare for logging.

PrepareOpenTracing collection.

PrepareMetrics::Any::Adapter to collect metrics.

Starts the main loop.

Applies signal handlers for TERM and QUIT, then starts the loop.

Microservices are hardly a new concept, and there's a lot of prior art out there.

Key features that we attempt to provide:

• reliable handling - requests and actions should be reliable by default
• atomic storage - being able to record something in storage as part of the same transaction as acknowledging a message
• flexible backends - support for various storage, RPC and subscription implementations, allowing for mix+match
• zero transport option - for testing and smaller deployments, you might want to run everything in a single process
• language-agnostic - implementations should be possible in languages other than Perl
• first-class Kubernetes support - k8s is not required, but when available we should play to its strengths
• minimal boilerplate - with an emphasis on rapid prototyping

These points tend to be incompatible with typical HTTP-based microservices frameworks, although this isoffered as one of the transport mechanisms (with some limitations).

Here are a list of the Perl microservice implementations that we're aware of:

• https://github.com/jmico/beekeeper - MQ-based (via STOMP), usingAnyEvent
• https://mojolicious.org - more of a web framework, but a popular one
• Async::Microservice -AnyEvent-based, using HTTP as a protocol, currently a minimal wrapper intended to be used with OpenAPI services

Although this is the textbook "enterprise-scale platform", Java naturally fits a microservice theme.

• Spring Boot - One of the frameworks that integrates wellwith the traditional Java ecosystem, depends on HTTP as a transport. Although there is no unified storage layer,database access is available through connectors.
• Micronaut - This framework has many integrations with industry-standardsolutions - SQL, MongoDB, Kafka, Redis, gRPC - and they have integration guides for cloud-native solutionssuch as AWS or GCP.
• DropWizard - A minimal framework that provides a RESTfulinterface and storage layer using Hibernate.
• Helidon - Oracle's open source attempt, provides support for two types oftransport and SQL access layer using standard Java's packages, built with cloud-native deployment in mind.

Most of Python's frameworks provide tools to facilitate building logic blocks behind APIs (Flask, Django ..etc).

For work distribution,Celery is commonly used as a task queue abstraction.

• https://rocket.rs/ - although this is a web framework, rather than a complete microservice system,it's reasonably popular for the request/response part of the equation
• https://actix.rs/ - another web framework, this time with a focus on the actor pattern

JS has many frameworks that help to implement the microservice architecture, some are:

• Moleculer - generally a full-featured, well-designed microservices framework, highly recommended
• Seneca

• Swoft - async support via Swoole's coroutines, HTTP/websockets based with additional support for Redis/database connection pooling and ORM

Microservice support at the provider level:

• AWS Lambda - trigger small containers based on logic, typically combinedwith other AWS services for data storage, message sending and other actions
• "Google App Engine" - Google's own attempt
• Heroku - Allow developers to build a microservices architecture based on the services they providelike the example they mentioned in thisblog

Deriv Group Services Ltd.DERIV@cpan.org

• Tom MolesworthTEAM@cpan.org
• Paul EvansPEVANS@cpan.org
• Eyad Arnabeh
• Nael Alolwani

Copyright Deriv Group Services Ltd 2020-2022. Licensed under the same terms as Perl itself.