Page tree
Skip to end of metadata
Go to start of metadata
DescriptionDevelop a distributed framework for management, monitoring and automation of deployment and usage at the local and community scope

Title: Development Dashboard, Monitoring and Automation Framework for Pharos Community lab to handle requests for resources from the OPNFV community at large.

Description: Develop a distributed framework for management, monitoring and automation of deployment and usage at the local and community scope (granularity)

  1. Draft framework for functional parts of a central dashboard application as well as work in the development of the specification of the framework for lab inter-connectivity and component parts
  2. Establish a framework for monitoring (dashboard functional component) that includes connectivity methodology (inter-site and intra-site).
  3. The booking tool will allow for:
    i) Different classes of booking times, resolution of booking times and exit criteria of "release" (fail, finished, expired booking time)
    ii) Enforcement, management, optimization of the resources. Including the balancing and re-balancing of HW resources between Dev and CI needs.
    iii) Class of user and capabilities
    iv) SSH to each server an have each node check for internet connectivity
  4. Additional components for design and integration include HW to Switch mapping algorithms (potential for a white-paper on Datacentre Operations here) and framework to allow for plugin / adapter based architecture.
  5. Integration to JIRA, Yardstick and other relevant reporting and administration / operations tools and frameworks
  6. Stretch Target - automation that resolve intra-DC setups (multi-site, Geo-red, geo-distributed application)

The project will proceed in the following manner, with the following subset tasks/deliverable

    • Pilot project (based on working prototype from J. Lausuch)                
      • Delivers:
        • Goal and Use Case description of the Pilot (input from groups needed -  see below)
        • Definition of Metric, Measurements and "how we want to represent" information within the tool 
        • Working implementation to the point of "Ready for Adoption" request to INFRA, Project and Program Group
        • Documented communication/process outline for Projects on "how they want/view" the booking of resources (expectation check)
    • Adoption Project (implementation and "live use")
      • Delivers:
        • A maintainable implementation and a adoption program (getting users on board)
        • A CR process (how to change the use, change the input/output) for the tool
        • A hosting location or method (single place (on LF) or distributed - each lab can have a 'local copy' of the overall picture - message bus implementations?)                 

Success metrics(to be defined in greater detail in Project/Infra group meeting):

  1. A working booking tool that satisfied the requirements set out in the Pilot
  2. Integration into the release project (Adoption by projects)
  3. A documented framework that allows for installers (and others) to have their code deployed,executed or in other ways "acted" upon in a community lab from a central (or centrally-distributed) tool.
  4. A reporting mechanism in place that allows for the measurements and metrics as outlined in the pilot
  5. Stretch - Batch Run of Select Release criteria from tool (i.e ordering up a tesing release across labs)

Additional Information: Pharos

Desirable Skills: LAMP skills, Python, Flask, 0MQ (or Rabbit / Messaging Bus tech), Hardware familiarity (ILOM, IPMI, iDRAC, UCS, etc) an asset

Expected Outcome: An automated Resource (Lab) booking, monitoring, deploying and reporting tool that is the central point for all resource requests from the OPNFV community to the OPNFV Infrastructure Community (lab providers)

Difficultly: LOW / MEDIUM /  HIGH
(This is a Priority Target for the Infra Working Group for 2016, as such the candidates will be part of a motivated team of Lab owners who's targets are to see this implemented)

Desired project timeline/completion date: Pilot completed for C-release (basic booking cases), automation and full features targeted for D-release

Mentor(s) & contact info:

Intern: Max Breitenfeldt