SNIC User Forum, Dec. 16 2014
“I've told you once.
No you haven't!
Yes I have.
When?
Just now.
No you didn't!
Yes I did!
You didn't!
I did!
You didn't!
I'm telling you, I did!
You did not!
Oh I'm sorry, is this a five minute argument, or the full half hour?”
*From the Monty Python argument sketch
The responsibilities of SNIC should start somewhere and end somewhere, the question is where these start and endpoints lie.
SNIC should service and help all research that requires compute and/or storage resources
SNIC should only focus on servicing research that requires large scale HPC and associated storage resources
SNIC should provide more compute capacity
SNIC should have fewer and larger national resources
Is geographical proximity important | Ratio |
---|---|
Yes | 14% |
No | 86% |
SNIC should not impose job size requirements on any compute resource
All SNIC compute resources should accept jobs with a run time of 3 days
SNIC should harmonize login procedure, job submission procedure, job scheduling behaviour, and software availability
SNIC should provide services for long−term storage of research data
SNIC should invest in improving the interfaces to SweStore
SNIC centres provide good support and documentation
Satisfaction level | Ratio |
---|---|
Poor | 0% |
Less than satisfactory | 1% |
Satisfactory | 30% |
More than satisfactory | 30% |
Excellent | 31% |
No opinion | 9% |
Support doesn't scale well. How should SNIC cope with getting more users?
Other infrastructures that use SNIC resources must handle end-user support for their community
More funding must be allocated for handling all kinds of support activities
SNIC application experts should focus on operational services regarding development environments and associated training
SNIC should provide resource allocations that run for longer than 12 months
Matches projects | Ratio |
---|---|
Yes | 40% |
No, I would prefer to have allocations with longer duration | 41% |
No answer/ Not applicable | 17% |
Other | 2% |
Large SNAC allocations should be integrated with VR grants