20180606- Panel Session: Project Governance & Trends Analysis

-Slide 2 add reference to TeraGrid

-Slide 3 replace org chart or use the one Ron sent

-Slide 10 what is the status of the Staff Climate Survey

-Slide 11 update names

-Slide 6 Suggestion from Shawn Strande via chat to include names of XAB and UAC and not just that they exist

-Slide 16 suggestion that it is a rabbit hole–remove it

-Slide 7 missing slide number

-Before slide 2 need an introduction to transition to new topic

-Slides 4 and 5 will raise questions; have a list of contributing factors:

• non-FTE budget as an explanation (explains NCSA)
• Unspent hardware investment; emergence of public cloud acceptance holding for that transition–will hear more in Ops discussion
  • add another color for hardware line
• Source of what the projected is
• Travel will increase with PEARC
• Projection is not based on reality
• Use annual for projection rather than 6 months
• Purposes for underspending
• Staff turnover – would be good to have data on that for hiring lag
• Year 1 partners were spending down carryover from prior award
• Need to then answer the question what are you going to do about it
• TACC explanation for lean amounts
• Authorization to spend in February, and award processing takes time
• Emphasize the % under spending rather than value and how that calculates into weeks (5 weeks?)
• Large award
• Illinois takes back unspent funds; whatever carryover allow to keep either 2 months average spend up to a max of $200k so some money is tied up in that carryover
• Managing responsibly

-Provided TACC projected amounts with explanation for lean amounts (Kelly/Karla)

-Use of term "Traditional users" is different in ECSS...logins, command-line users, etc.; send Nancy graphic

-R1 terminology now called Major Research Institution (MRI)

-Visualization isn't going down, but you're seeing heterogeneous systems where the work is going

-Look at growth of GPU users

-Caution: classifying by resource type not by the work they are doing; not sure it will help case. Data-intensive computing portion is wrong–remove or refine

-Trend in usage of GPUs and Phis

-Trend of diversity from Nancy's slide

-How are people solving the problems today and how is that changing?

-Talking narrowly for those who have allocation on XSEDE, there could be extended community............use of users could be defined

-Fields of science trends

-Keywords query of abstracts

-Trends for L2s to consider–should be talking about these things–qualitative or quantitative data observations

• • Does XCI campus interaction equate to institutions deploying?
  • XCI Instrumenting software components to determine usage?
  • CEE champions, trainings, student preparation, PI interactions, etc.