Jump to: navigation, search


< Blazar
Revision as of 12:44, 11 July 2014 by Pafuent (talk | contribs)

Project codename


Current incubation requirements status



Resource reservation service for OpenStack based clouds. Blazar provides leasing opportunity to manage different cloud resources in time-based manner.

Mission statement

Program: Resource Reservation

Mission: Provide ability to manage different cloud resources in time, by offering their planning and resources allocation/reclaiming process automation.

Detailed Description

Blazar provides users and cloud providers opportunity to use resource reservation in their OpenStack clouds. Primary idea of creating Blazar originated with two different use cases:

  • Compute host reservation (when user with admin privileges can reserve hardware resources that are dedicated to the sole use of a tenant)
  • Virtual machine (instance) reservation (when user may ask reservation service to provide him working VM not necessary now, but also in the future)

Now these ideas have been transformed to more general view: with Blazar users can request the resources of cloud environment to be provided (“leased”) to their project for a specific amount on time, immediately or in future.

Blazar operates reservations as basic units in its logic. Reservation is an allocation of certain cloud resource (Nova instance, Cinder volume, compute host, etc.) to particular project. Reserved resource is taken from common resource pool depending on what kind this resource is - for example, reserved VM will be in shelved state till it is not needed by user. Both virtual (Instances, Volumes, Networks) and hardware (full hosts with specific characteristics of RAM, CPU and etc) resources can be allocated via '"lease'".

Lease is simply some kind of a negotiation agreement between the provider (Blazar, using OpenStack resources) and the consumer (user) where the former agrees to make some kind of resources (both virtual and physical) available to latter, based on a set of lease terms presented by the consumer. Here lease may be described as contract between user and reservation service about cloud resources to be provided right now or later. Technically speaking, lease is a group of reservations granted to particular project upon request. Lease is characterized by start time, end time, set of individual reservations and associated events.

Event is simply something that may happen to lease. In most simple case event might describe lease start and lease end. Also it might be notification to user (e.g. about soon lease expiration) and some extra actions.

Now Blazar is tightly integrated with core OpenStack projects, such as Nova and Keystone (including usage of trusts mechanism) and is using common OpenStack practices for projects building: Blazar repositories obey tox, pbr and global-requirements. Blazar has full technical i18n support and uses OpenStack common code where it is appropriate (including oslo.config and oslo.messaging support). There is current review under progress for using Pecan+WSME to implement REST API and there is Devstack integration implemented. Now this Reservation-as-a-Service initiative includes three Stackforge projects:

  • Blazar itself
  • Blazar client (with implemented comfortable CLI)
  • Blazar-Nova project (that includes exclusively Nova-related changes such as filters and API extensions for Nova)

Blazar implements pluggable architecture (using stevedore) to support different resources to be reserved from different OpenStack projects. Virtual instances and compute hosts support is already implemented, in future Blazar is going to provide volumes, network resources and storage nodes reservation, that's why we consider it as a completely new OpenStack program Resource Reservation.

Blazar Roadmap

Blazar Roadmap

Location of project source code

Programming language, required technology dependencies




All Blazar dependencies match OpenStack global requirements and use only common OpenStack dependencies:

Is project currently open sourced? What license?

All Blazar initiative projects are open sourced and use Apache 2.0 license.

Level of maturity of software and team

Blazar initiative has been engaged last year, and now there is great team working on including talented people from Mirantis, Bull, Inria, Intel, eNovance and Rackspace. We believe that this project will have great future as a part of OpenStack ecosystem and will bring new features and possibilities to it.

Proposed project technical lead and qualifications

Dina Belova, who is Software Engineer at Mirantis, (DinaBelova on IRC) is the current Blazar PTL. Her main responsibilities are architecture design and community-related work in Blazar (including holding weekly IRC team meeting, preparing agenda for them, managing Blazar Launchpad). She is top Blazar reviewer and one of the top Blazar contributors, and she is working with OS ecosystem more than two years, creating private OpenStack based clouds for Mirantis customers and contributing to official OpenStack projects. She's been elected by Blazar community due to common OpenStack practice - https://wiki.openstack.org/wiki/Blazar/PTL

Other project developers and qualifications

Core team

In addition to Dina Belova:

  • Sylvain Bauza (bauzas on IRC) - is OpenStack Software Engineer at Bull with more than 10 years of experience in designing, managing and operating large Web farms. He is now in charge of adding value to the XLCloud project (http://www.xlcloud.org) by operating and acting on its Openstack stack and thus contributing to Blazar for delivering a reservation service.
  • Nikolay Starodubtsev (Nikolay_St on IRC) - is Software Engineer at Mirantis. He is working in OS ecosystem more than two years and has much experience in implementing customer-specific features. He took significant part in implementing core Blazar possibilities and has been working on this initiative from very beginning of Blazar.
  • Swann Croiset (scroiset on IRC) - is OpenStack Software Engineer at Bull since last year. He has taken part in Blazar developpement since the very begining to provide reservation service for the XLCloud project. He will work to provide integration tests and stress tests to the project for next release, since he has 5 years performance and integration testing experience before landing on the planet OpenStack. His other assignments are also the billing and monitoring, leveraging Telemetry program.
  • Pablo Andres Fuente (pafuent on IRC) - Software Engineer at Intel. He's working on development of the virtual reservations idea, and his OpenStack experience includes much work with Orchestration Program.

Active code contributors

  • Sergey Lukjanov (SergeyLukjanov on IRC) - Senior Software Engineer at Mirantis, OpenStack Data Processing program PTL (Savanna project) and one of the core-team members in OpenStack Infrastructure program. He is taking part in Blazar architecture design and its development since very beginning and is mentoring its development and community related questions.
  • François Rossigneux (f_rossigneux on IRC) - Software Engineer at Inria. Is working on implementation of physical reservations use case for XLcloud project, and has great view on how it should work with Blazar core features.
  • Yuriy Taraday (YorikSar on IRC) - Senior Software Engineer at Mirantis, who is taking active part in OpenStack development during last three years. He has great experience in all OpenStack core projects, and helps much with Blazar integration in entire OS ecosystem.
  • Christian Martinez (chmartinez on IRC) - Semi Senior Software Engineer at Intel. He began his Blazar implementation experience with tests functionality improvement, but his entire interest lays near virtual reservations, so he's working on its better functionality.
  • Dmitriy Dyachkov - our only independent developer :) He started his OpenStack experience with our project, and we hope that his OS journey will be nice and comfortable :)

And also special thank to people, who've been taken part in Blazar, although it's not their main point of interest:

  • Julien Danjou (jd__ on IRC) - OpenStack Telemetry program PTL, he inspired Blazar to be born, and he is mentoring our project to fit all OpenStack practices.
  • Michael Still (mikal on IRC) - Software Development Manager at Rackspace
  • Nikita Konovalov (NikitaKonovalov on IRC) - Software Engineer at Mirantis

Naming issues

Blazar (formely know as Climate) team searched for the new name because it turned out that 'climate' repos already exist on PyPi and readthedocs.

Infrastructure requirements (testing, etc.)

All our code/reviews and bugs/blueprints are hosted at OpenStack Gerrit and Launchpad correspondingly. We have unit tests and all flake8/hacking checks runnong at OpenStack Jenkins and we hope that we’ll implement integration tests in the OpenStack infrastructure system (using Tempest) in nearest future. Currently all documentation is published on OpenStack Wiki (http://wiki.openstack.org/wiki/Blazar). When the renaming from Climate to Blazar is finished all documentation will be moved to readthedocs.

Have all current contributors agreed to OpenStack CLA?


Related links