Discussion - Matt Crawford and Shawn McKee, scribe: Les Cottrell
Problem Topics:
* Path allocation:
o
Contributors: OSCARS, Terapaths, BRUW, USN, UL,, VINCI, DRAGON
o
Users: Lambda station, MonALISA
*
Monitoring: Diagnosis, pS, MonALISA
o
Must include network, hosts & application
*
Problem space: intra-US, extra US, application integration, scheduling deadline and time share
*
Network resources to pillage & exploit: USLHC, Abilene, ESnet/SDN, GEANT (USN UL GLIF)
*
Tests, goals. milestones:
o
max T0->T1 with competing traffic, does it work stably, does it perform,
o
T1<=>T2 exercises soar with the winners, individually and all together
o
Need goals of when these demonstration tasks (Exemplars) are completed. Agreed to by April 07, referred to as Network Circuit Challenge (NCC). N.b. ideally this requires the end sites to be at 10Gbps. Not all ATLAS T2 are at 10Gbps so will target rates of 1Gbps, 3Gbps and 8Gbps. The main emphasis is for T2 (not T0->T1). Need to identify people to lead this effort, by default it is Shawn McKee & Matt Crawford.
* Craig proposed that Shawn Mckee (USATLAS) and Matt Crawford (USCMS) will assemble a plan, setup task, scope and date for Tier 1<=> Tiet 2 network circuit challenges. Matt will collect the project inventory from Brian Tierney.