[CIG-MC] Request for Feedback: Geodynamics in a Cloud

Eric Heien emheien at ucdavis.edu
Wed Jun 6 09:03:12 PDT 2012


Hello,

I'd like to thank everyone who took the time to respond to the survey, your answers and comments are extremely helpful.

For your information here are the totals for the 1st and 2nd questions.  Total responses are indicated in the brackets.

1) I or my students would be interested in using prepackaged virtual machines (VMs) of CIG software.
[11] Yes
[ 6 ] No
[ 1 ] Don't know

2a) If you answered "Yes" to 1, please tell us why you are interested (select 1 or more):
[ 5 ] So that I won't need to upgrade my system
[ 8 ] So that I won't need to compile source or install extra libraries
[ 7 ] Because my simulation usage fits cloud computing (i.e. occasional large runs)

2b) If you answered "No" to 1, please tell us why you are not interested (select 1 or more):
[ 4 ] Everything already works fine, I see no need for more complexity
[ 0 ] I don't like/trust cloud computing
[ 3 ] I already have a computing allocation and don't need more
[ 2 ] I'm concerned about code performance in cloud computing
[ 3 ] My grants/institution won't pay for cloud computing

2c) If you answered "Don't know" to 1, please tell us more:
[ 0 ] I've never heard of cloud computing or VMs
[ 0 ] I haven't had a chance to use any CIG codes
[ 0 ] I'm unsure about how hard it will be to use
[ 1 ] I'm unsure about the performance on a cloud

Many of those who responded "no" indicated they were happy with how things work now and didn't want to add complexity.  To make it clear, CIG would not be offering the VMs as a replacement to current source/binary distributions but rather an additional way of running the code.

CIG will begin testing the waters by preparing a virtual machine for tutorials in the upcoming CIDER summer program.  Based on the response there we will decide how to proceed.

Again, thank you all very much for your input.  Feel free to let me know if you have any additional comments or questions.

-Eric


More information about the CIG-MC mailing list