LHCONE R&D and RNT-WG call

Tuesday 12 Sept 2023, 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 Preparation of SC23 demos and DC24 mini-challenges

Present: Edoardo Martelli, Harvey Newman, Bruno Hoeft, Phil Demar, Dale Carder, Syed Asif, Tristan Sullivan, Yatish Kumar, Justas Balcas, Joe Mambretti, Bill Johnston, Marian Babik, Tim Chown, Raul Lopes, Tom Lehman

 

Bill map:
- The map files (.pptx, .jpg, and .pdf) are at: https://www.dropbox.com/sh/p2wcyypvubeki7q/AAAMJGF5O9sxUFospm3pRLzLa?dl=0
 

JoeM on SC23
- Thomas Tam is making progress with the 400G link between Triumf (UVic-SFU) and the Pacific Wave at the Seattle Pacific Northwest GigaPoP
- The 400 Gpbs connection between Pacific Wave at Seattle and StarLight is also progressing (It will be provisioned as a prototype of AP-REX 2.0 soon to be renamed NA-REX)
- SCinet has implemented 1.2Tb between the Joint Big Data Testbed Facility at McLean, Virginia and Starlight. This path is being used for 400 Gbps flow testing between those sites. This 1.2 Tbps path will be extended to the StarLight booth at the SC23 venue in Denver. The booth will have 5*400 Gbps paths connecting it, as well as multiple 400 Gbps DTNs
- SCinet WAN team is doing the mapping of the NREs to VLANs and to the specific circuits they will use (this task should be completed by the end of the week)
- We have not discussed the scheduling of the demos, yet. We have to do this later

Yatish about High Touch:
- ESnet is prioritizing the deployment in Denver and at StarLight. The implementation is progressing well. They have to measure production ESnet traffic (e.g., they cannot use the ESnet testbed)
- Where are the DTNs and who can be contacted to check the DTNs are correctly connected?
   - One will be at UVIC, sending data to the DTN  at the SC show floor.
   - High Touch can measure this traffic at StarLight and in Denver.
   - Marian and Edoardo will work on the DTNs at CERN
   - there is currently no visualization tool. Something should/could be developed.
   -  Bruno: at KIT there will be two Power9 DTNs. Two more Intel DTNs may become available, TBC

NOTED and other demos
- FNAL will have a dedicated router for the SC demos (part of the new FNAL network testbed). Phil will organize a meeting to define what is needed there.
- The documentation produced by Tom is here https://drive.google.com/drive/folders/1AxhOaGgVvxgSkhb0KGDSMhE-S_oOU-38

17:20 → 17:40 RNTWG

CHEP2023 paper deadline is in 10 days, many thanks to all who contributed. 

https://www.overleaf.com/read/fcpnpxtjvjby

Collectors and dashboards meetings:
- Notes are at https://docs.google.com/document/d/1vwOVJ4BLp_oUM-mdSkeIFG20A-0Q37FFhzwcbTqujK0/edit?usp=sharing
- Reviewed the current status of the network of receivers and discussed plans for SC23
- Discussed how we could visualize our data (and what data we have available)
- We're planning to explore the following directions: 
   - Possibilty to add netlink data to UDP fireflies (either as updates or as summaries in end state) - can be useful also for jumbo frames and congestion algorithm studies
.  - Plot UDP fireflies alongside network traffic (utilisation)
.  - Check if we could correlate with WLCG site network monitoring information (or DDM/FTS) 

FTS gfal2 implementation is progressing.
dCache testing has been restarted following re-deployment of the next version (we have also received source code, so it should be easier to follow up).

17:40 → 17:55 AOB

NtR

17:55 → 18:00 Next meeting

Next meeting in two weeks: Tue 26 of September at 17:00 CEST