You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 10 Next »

Meeting Minutes from Performance Working Group at 2012 Joint Techs in Baton Rouge

Monday, Jan. 22, 2012

http://events.internet2.edu/2012/jt-loni/agenda.cfm?go=session&id=10002152&event=1223

Internet2 Update (Jeff Boote)

- Recently ported OSCARS v6, which is an IDC protocol ready for initial consumption that has been ported to OpenFlow
- Currently evaluating priorities and roadmap for perfSONAR

The Challenge (Carla Hunt and Ken Miller)

See Carla's Slides

Question:In your presentation on the performance portal this morning, how are you planning to capture flow data?
Answer:We have been evaluating tools internally. So far, NMON, looks promising. We are still working on the database schema.

Community comments following Carla and Ken's presentations:

- There was some agreement on usefulness of being able to specify MTU size.

- Also a comment that it would be nice if the perfSONAR interface reflected results of MTU related to iPerf/BWCTL.
- Is there any way to co-schedule throughput and latency measurement tests on perfSONAR toolkit nodes?
- Ability to centralize storage of test results?
- Application layer testing
- Current toolkit provides a gui to configure tests on one host.

- A bit more challenging to develop a gui way to configure a constellation of scheduled measurement tests.
- How can a NOC check to see if Multicast is working?

- How much Multicast traffic is really being generated?

-NLANR Multicast beacon used to be useful for this.

-Possible the project has not been updated in a while.

Outreach to Research and Campus - SBIR program (Rich Carlson)

  • Rich Carlson talked about the DOE SBIR program in the Office of Science.
  • He is looking for better tools to run performance tests and manage infrastructure.
  • Think about partnering with or starting up a new business. 1.1 M of funding over three years.
  • More information: http://science.energy.gov/SBIR/

We voted Ken Miller in officially as new co-chair and discussed the classification of the group. Should it be classified as a working group or special interest group? Still an open question.

  • No labels