LHCONE R&D and RNT-WG call

Tuesday 27 Aug 2024, 17:00 → 18:00 Europe/Zurich

Edoardo Martelli (CERN), Joe Mambretti (International Center for Advanced Internet Research Northwestern University), Marian Babik (CERN), Shawn Mc Kee (University of Michigan (US))

17:00 → 17:20 LHCONE

Presents: Joe Mambretti, Marian Babik, Edoardo Martelli, Carmen Misa Moreira. Bill Johnnston, Jim Chen, Trustan Sullivan, Doug Southworth, Harvey Newman, Hiro, Tom Lehman, Chris Walker, Tim Chown

SC24
- The UCSD Gen4 server has arrived to CERN, but it doesn't start. CERN and UCSD and looking how to fix it.
- Harvey: Caltech is planning to have 6x100G connection to the booth. But the cost is probably too high (each fibre cost 5000$)
- Tom can start working on the VLANs
- Bill has issued a new version of the SC24 map.
- Starlight is working with CANARIE for the 400G connection from Chicago to UVic
- Edoardo will ask SURF if possible to upgrade to 400G connection CERN to Netherlight in Amsterdam

 

AOB
CMS and ATLAS would like to delay the start of Long shutdown and another delay of 6month of the start of Run4. Which would start in 2030.

 

17:20 → 17:50 RNTWG

LHCOPN/LHCONE workshop
- There will be a dedicated session on Scitags on Thursday 10th of October afternoon (Beijing time)
- Ideas for contributions are welcome 

SC24 - Starlight team has successfully re-run the SC23 demo

CERN RCS-IT Scitags project has been approved
- Plan to meet with the experiments to gather their requirements and ideas 

IETF RFC 
- Draft updated and presented at IETF 120 
- Received additional feedback to clarify limited domain, add some RFC references and describe our experiences with HbH (DO)

Collectors
- New PR: https://github.com/scitags/firefly-collector/pull/1
- Missing Kafka input/output plugin (A: Andy Lake)

Flowd
- New release - 1.1.4 
    - Added experimental plugins developed for SC23 (following up on feedback from Starlight team)

Storages
- Wrote up a proposal for some minor changes in how storages should process HTTP headers and issue fireflies
    - Main change is to ensure that fireflies content is always in the direction of the data transfer (src is always the node sending the data)
    - Clarifies handling of the HTTP headers and their forwarding 

17:50 → 17:55 AOB

17:55 → 18:00 Next meeting

Next meeting: Tuesday 10th of September