|Topics||Time budget (Max 2 hours)||Participants||Session Type (e.g. discussion, hacking, office hour, presentation, etc.)||Notes||HW resource plan/requirement|
|Best Practices/Use cases|
Telco view on acceleration and requirements of VNFs.
|Edge cloud: requirement and architecture design for China Mobile||1hr||presentation||This presentation will introduce the edge cloud requirement, architecture design, deployment plan and issues we still have within China Mobile. I will also explain how open source communities can help end users in this effort.|
|Release Automation||1hr||presentation (15m), discussion||Discuss release automation status and collection requirements/plan work for Gambia release.|
|OPNFV Verified Program/Dovetail||1 hour||office hour: high level presentation (30min) + Q/A and discussion (30min)|
|Evaluation of Benchmark Repeatability , aided by new methods.||1 hour at start of Plugfest, plus many status breakout sessions, and 1 hour near end to sum-up||Discussion, Hacking, Testing, Results review|
The purpose of this testing is to evaluate result repeatability across two dimensions, Testers and Pods. Also, a comparison of results with existing methods and new methods would be valuable to TST009 development.
Can IDF and PDF help to explain the differences in results?
|Probably one or more remote Pods.|
|Monitoring System Behaviors under Long Duration Testing||We can have a 2 hours session to introduce and start the test, then show/discuss the monitoring board and system behavior during several days||Discussion, Hacking, Testing, Results review||Since we have long duration test POD ready and several test cases are also developed, we need to see/share the results and discuss about the gaps of SUTs, tests and monitoring, to make the whole pack of LDT suite be more mature.||Nokia Pharos POD|
|NFVi and VNF characterization using Yardstick/NSB||2hrs|
Presentation, Discussion, Demo
Tutorial on Yardstick/NSB:
|Minimum test requirements for CD release process gating||2 hrs||Discussion||Develop a proposal for minimum test requirements for CD release process gating.|
|Project team hacking/discussions|
|edge cloud project hacking||1h|
|opnfvdocs||2 hrs||Sofia Wallin||docs hacking, discussions, G release planning and scope|
|Yardstick||2 hrs||Feedback, discussions and G release planning|
|DP Benchmarking with TRex||2 hrs (will need several meetings)||discussion|
Review and compare how various DP benchmarking tools based on TRex are using the TRex API to perform their DP benchmarking:
This will be first step towards a more standardized way to do NFV DP benchmarking across tools with results that can be compared.
Note that the findings from these discussions will also be applicable to other traffic generators.
OPNFV Projects that are using TRex:
VSPERF, Yardstick, NFVbench,
External projects using Trex:
redhat tool (name to be provided by redhat), FD.io CSIT
|DP testing with Bottleneck||1h||discussion/hacking||Review how NFVbench can be integrated into Bottleneck and how to combine with other tools to generate mixed workload (data plane/control plane/storage plane)|
|Other/Wild Card topics (e.g. community, new project proposal, etc.)|
|OPNFV release process||2 hrs||discussion|
|Fraser release process retrospective||2 hrs||David McBride||presentation and discussion|
|OPNFV Hands-on program||1h||discussion||Kick off discussion of OPNFV hands-on program. Quick start and lecture documents of OPNFV are very useful while getting new developers on boarded to OPNFV work in community and even on their own as an OPNFV user. Such things can be used in a tutorial session in various events, especially workshop events in local user groups. We'll share materials which were used in OPNFV Hands-on event in Tokyo first, then discuss better way to have such program in community wide.|
|Workgroup report outs||1h||presentation/discussion||For Edge & OVP|