Attending: John B, Marcus, Matt, Jens (c+m), John H, Winnie, Daniel, Luke 0. Operational blog posts Still need more blog posts! No operational issues. 1. The JISC end-to-end workshop : some presentations available. https://www.jisc.ac.uk/events/campus-network-engineering-for-data-intensive-science-workshop-19-oct-2016#resources Lydia presented work with GridPP - DiRAC/GridPP data transfers from Durham (and, eventually other DiRAC sites) to RAL; other interesting work includes the Science DMZ 2. Remaining loose ends (the ones mostly not covered last week) - current state and any $\Delta$ * Accounting (no news) * Information systems and ATLAS (no news) * use (and status) of test SEs (?) * data node writeup and Globus endpoints and endpoint implementations and endpoint implementations testing (no news) * IPv6 and data transfers (and test transfers)(?) * Anything for the cloud-at-Crick workshop? (linking GridPP to cloud?) * Science DMZ/JISC campus/eInfra (see above) * GridPP as a data einfra - comparing to climate, "UKT0", SKA Other stuff: Here's the ECFA workshop Daniel mentioned a few weeks ago - https://indico.cern.ch/event/524795/timetable/ There is a GDB coming up (next week) - http://indico.cern.ch/event/394788/ Two things may (or may not) be storage related: 'performance' at WLCG [workshop], and 'regional federations demonstrator' There's the DPM workshop of course: http://indico.cern.ch/event/559673/ We spent some time talking about high level bigpicturey stuff, like GridPP and infrastructures and UKT0 and that sort of stuff; a bit hard to summarise but sometimes it could be useful to try to set the big picture and how GridPP fits into it, or could fit into it. 3. AOB Jens Jensen: (02/11/2016 10:02:25) https://www.jisc.ac.uk/events/campus-network-engineering-for-data-intensive-science-workshop-19-oct-2016#resources Daniel Peter Traynor: (10:09 AM) QM has test SE s storm ipv6 and version testing also have a test xrootd standalone server for the same perpose Marcus Ebert: (10:10 AM) ECDF has a test SE for testing new DPM versions, only used by Atlas for testing Daniel Peter Traynor: (10:30 AM) mtu discovery? icmp it is perfsonar latency