Network and Transfer Metrics WG - Metrics Area Meeting

Europe/Zurich
28/R-015 (CERN)

28/R-015

CERN

15
Show room on map
Marian Babik (CERN), Shawn McKee (University of Michigan ATLAS Group)
Description

The meeting date/time is a result of http://doodle.com/ezrfh8eybu7iybxy

Details on the Network and Transfer Metrics WG are available at our Twiki.  

Slides

Network and Transfer Metrics WG, Metrics Area Meeting (26 Nov 2014)

Attending: Laurent Caillat-Vallet, Costin Grigoras, Ilija Vukotic, Frederique Chollet, Duncan Rand, Kaushik De, Hung Tee Lee, Jorge Alberto Diaz Cruz, Enrico Mazzoni, Bruno Hoeft, Shawn McKee
at CERN: Tony Wildish, John Shade, Michail Salichos, Marian Babik

Indico: https://indico.cern.ch/event/354593/

Next meetings:

– Fixed schedule for 2015
• 28 Jan, 18 Feb, 18 March, 8 Apr (all at 4pm CEST)

 

List of actions:

- Use cases and requirements document (https://docs.google.com/document/d/1ceiNlTUJCwSuOuvbEHZnZp0XkWkwdkPQTQic0VbH1mc/edit)

Please review/add your input:

  • FTS, FAX, PhEDEx, Panda, Rucio, by 5th of Dec 
  • Experiments by 12 of Dec (Marian to ping ATLAS, LHCb)

- XRootD monitoring status, Marian to organize meeting on GLED status with Julia, Ilija, Mateusz, Shawn

Minutes:

Status of operations and comissioning of perfSONAR network was presented by Marian.

Kaushik: What would be good time to take a look at the real data ?

Marian: Since we're in a middle of the update campaign, it's better to wait until most of the sites are updated (current deadline for sites is 8th of January). The current ITB data store already contains data and can be queried, but we need more sites to udpate to have more complete data sets.

Kaushik: would it possible to get the data via SSB/AGIS as before ?

Shawn: In principle yes, but we will need to update the scripts as sites are running different measrument archives now, to be followed up with ATLAS.

Shawn commented on IPv6. It will be enabled and when sites have both IPv4 and IPv6 TWO tests will be run automatically.  The impact we need to consider is that the number of tests will increase as we have more sites with IPv6.   Original manual mesh forced only IPv4.  Not the case in the new mesh.

perfSONAR metrics, examples of existing measurements and examples of identifying potential network issues was presented by Shawn.

John commented that it would be great to document this in order to encourge sites to see how usefull the tool is.

AOB

Shawn will be at CERN on Thursday 4th December, so we will have perfSONAR office (28-R-014) during the morning (until ~11:30), feel free to join us if you'd like to discuss anything related to perfSONAR

 

 

There are minutes attached to this event. Show them.
The agenda of this meeting is empty