Jump to: navigation, search

Zaqar

Revision as of 14:51, 20 March 2014 by Kgriffs (talk | contribs) (Summary)

Mission

To produce an OpenStack messaging API and service that affords a variety of distributed application messaging patterns in an efficient, scalable and highly-available manner, and to create and maintain associated Python libraries and documentation.

Summary

Marconi provides a messaging platform for web and mobile application developers, supporting publish-subcribe, producer-consumer, point-to-point, and hyrbrid communication patterns. The project provides a simple, easy-to-use API designed according to direct feedback from users, and allows for a variety of options for backend storage, allowing operators to tailor the performance, durability, and scalability of their offerings to achieve a best-fit messaging solution for themselves and their users.

Detailed Description

In order to support more complex web applications running on OpenStack, a messaging service was needed. To fill this need, the Marconi project was proposed at the Grizzly design summit. Requirements were discussed with the community and used to form the basis for the project's charter. Implementation began in February 2013, and we have been fortunate to receive regular contributions from Red Hat, Rackspace, IBM, and others since that time.

Marconi's overarching goal is to provide web-scale, highly-available messaging to web applications that run on OpenStack. Marconi runs on Nova servers, behind OpenStack load balancers and uses Keystone authentication middleware. The Marconi implementation makes use of Oslo, and follows the standard OpenStack hacking guidelines.

Marconi provides an interface for posting messages, and later claiming those messages for processing. It also provides an interface to clients for listing messages without needing to claim them (ala RSS and Atom), in order to support pub-sub and passive auditing of producer-consumer workflows. The service guarantees first in, first out (FIFO) order for single producer models, best-effort ordering otherwise.

The Marconi architecture is pluggable in terms of both transport and storage. Reference drivers for HTTP (WSGI), SQLAlchemy, and MongoDB will be provided in the initial release (Icehouse), along with a SQLite driver to facilitate development and testing. Other transport and storage drivers have been proposed, and are currently under discussion. Marconi deployments will support HA, and will be able to scale horizontally in the transport and storage layers to enable large deployments. A routing proxy and migration service is also under development, to provide further horizontal scaling across multiple independent Marconi partitions, for use in extremely large deployments.

The Marconi v1.0 API defines the following operations:

API:

Get JSON home document
Check node health

Queues:

Create a queue
List queues
Set queue metadata
Get queue metadata
Get queue stats
Delete a queue

Messages

Post one or more messages
List messages
Get a message
Delete a message
Get multiple messages
Delete multiple messages

Claims

Claim messages
Query a claim
Update a claim
Release a claim

Core Team

Kurt Griffiths (kgriffs) is the current PTL. Other core team members include the following individuals who have demonstrated solid mentoring skills and good judgment while contributing to the program.

  • Flavio Percoco (flaper87)
  • Alejandro Cabrera (cpp-cabrera)
  • Fei Long Wang (flwang)
  • [your name could be here!]

Design

Marconi aims to be pragmatic, building upon the real-world experiences of teams who have solid track records running and supporting web-scale message queueing systems. The project's overarching design philosophy is derived from Donald A. Norman:

The value of a well-designed object is when it has such a rich set of affordances that the people who use it can do things with it that the designer never imagined.

Goals related to the above:

  1. Emergent functionality, utility
  2. Modular, pluggable code base
  3. REST architectural style


Principles to live by:

  1. DRY
  2. YAGNI
  3. KISS

Architecture

Marconi-architecture.png

Use Cases

  1. Distribute tasks among multiple workers (transactional job queues)
  2. Forward events to data collectors (transactional event queues)
  3. Publish events to any number of subscribers (pub-sub)
  4. Send commands to one or more agents (point-to-point or pub-sub)
  5. Request an action or get information from an agent (RPC)

Out of Scope

Marconi may be used as the foundation for other services to support the following use cases, but will not support them directly within its code base.

  1. Forwarding notifications to email, SMS, Twitter, etc.
  2. Forwarding notifications to web hooks
  3. Forwarding notifications to APNS, GCM, etc.
  4. Scheduling-as-a-service
  5. Metering usage

Etherpads

Presentations

  • Alejandro Cabrera. Rackspace Atlanta. Introducing Openstack Marconi. July 17, 2013. Youtube Speaker Deck
  • Flavio Percoco. EuroPython 2013. Marconi: Queuing and Notification Service for Openstack. July 2, 2013. YouTube
  • Kurt Griffiths, Allan Metts. Openstack Summit April 2013. Project Overview: OpenStack Queuing and Notification Service ("Marconi""). April 2013. YouTube
  • Kurt Griffiths, Flavio Percoco, Allan Metts. Openstack Summit November 2013. Openstack Queuing and Notification Service Marconi. November 2013. YouTube

Articles

FAQ

Will Marconi work with AMQP?

  • Planned as a backend for v2 API
  • Transport TBD
  • Talk to us about use cases
  • Need contributors

How does Marconi compare to AWS (SQS/SNS)?

  • Targets similar workloads
  • Marconi will provide a unified API to handle notifications and queuing
  • Marconi is highly customizable
  • FIFO and once-and-only-once guaranteed (depending on storage backend)

How mature is the project?

  • Marconi is used in production: marconi_users
  • Number of contributors is growing

What's next for marconi?

  • Releasing API v1.1 for Icehouse
  • Additional storage drivers
  • Storage sharding: scaling horizontally and heterogeneous storage
  • Notifications
  • Message signing
  • Additional ops features
  • Reference client library


See also the Icehouse Roadmap.

How easy is it to contribute/get up and running?

  • You don't need devstack
  • You decide what you want to work on: storage, transport, client-side
  • Very decoupled
  • Easy: choose a bug and submit a patch

Resources

Meetings Meetings/Marconi
IRC #openstack-marconi on Freenode
Trello https://trello.com/b/7NLODgbr (deprecated in favor of Launchpad)
Havana Spec Marconi/specs/havana
API Blueprint Marconi/specs/api/v1

Marconi/specs/zmq/api/v1
Sharding

Other Blueprints https://blueprints.launchpad.net/marconi
Milestones https://launchpad.net/marconi/+milestones
Developer Docs Tips for Developers
Python Client Bindings
Drivers
Guarantees
Incubation Marconi/Incubation
Source code https://github.com/openstack/marconi
Client source code https://github.com/openstack/python-marconiclient
Code Reviews https://review.openstack.org/#/q/status:open+project:stackforge/marconi,n,z
Bug tracker https://bugs.launchpad.net/marconi