is this a way we can compare the AWS server option and our personal computer and estimate the time needs to solve that model on the AWS server? Before starting the calculation we can estimate time needs and price?
1 Like
Hello Maziyar,
I guess it would depend on the VM’s that you use with AWS. I can’t really make any exact comparisons but there is some information in our docs that may be of use -
http://docs.itascacg.com/flac3d700/common/docproject/source/manual/program_guide/setup/aws.html?highlight=aws
Scroll to the bottom of the page and it lists FLAC3D speeds and pricing for various VM’s.
1 Like
Also have a look at this page: ITASCA Speed Test (itascacloud.com)
It’s a bit of a mess right now, but all of the computers that have names starting with EC2 are AWS machines. You could run the test on your own computer and on your AWS computer and the results will appear on this page.
2 Likes
Can you tell me how can I interpret the benchmarking results?
For example, meaning of the 4e6MCss, 4e6MCls…etc.
Regards.
I think 4e6MCss is a 4 million zone model using Mohr-coulomb constitutive model, run in small strain. 4e6MCls is the same model run in large strain. Subi refers to the Strain-softening ubiquitous joint model.
Thx. jhazzard.
I have found clues from the testing Python script below 