grawk Posted March 5, 2009 Report Posted March 5, 2009 I'm actually sort of surprised by this grawk. Is the model really that data intensive that disk access is a problem or do the researchers do something stupid like write every time step? I mean if you're writing to disk you're screwed as far at simulation throughput. One of the models is IO bound, and one is cpu bound. And all the post processing is IO bound.
Dreadhead Posted March 5, 2009 Report Posted March 5, 2009 One of the models is IO bound, and one is cpu bound. And all the post processing is IO bound. Oh ok. I wonder what's in the model to make it IO bound. Sounds like lots of look-up tables or something. I wasn't thinking about post processing , that's almost always io bound as you say.
grawk Posted March 5, 2009 Report Posted March 5, 2009 lots of weather modelling is based on lookup tables. The fudge factor for the "uh, this is what we think it will do"
Dreadhead Posted March 5, 2009 Report Posted March 5, 2009 lots of weather modelling is based on lookup tables. The fudge factor for the "uh, this is what we think it will do" Yup that's part of what makes me worry
grawk Posted March 5, 2009 Report Posted March 5, 2009 Why worry, they're accurate out to 4 hrs. Climate research is all a sham for funding anyway.
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now