Re: Introductions -- Re: Tungsten request (fwd)

[Frederick Gray <fegray@berkeley.edu>]

> I discussed the Loomis connection with our system manager. Currently
> Physics only connects with 100T to the core network. As long as we
> don't have a GigE connection, we would need 12 days to tranfer the
> 5 TB, we believe. It will change in the future, but currently the
> old fashioned way of moving LTOs seems our only option?
> 
> I will be going to PSI in a week from now and discuss their network
> situation with the experts there. Could you tell me which high speed
> networks NCSA in on, then I could ask more specifically.

Hi, Peter,

It is my understanding that PSI is on the SWITCH network.  The
unfortunate bit is that, according to this page:
  http://ait.web.psi.ch/services/costs.html
PSI is charged 20 CHF/GB peak and 5 CHF/GB off-peak for traffic 
to the USA.  A 5 TB transfer, even off-peak, would therefore cost
PSI 25 kCHF = $20k, which I think you will agree is not a small
amount of money. 

> If I am the user of the MuCap project, that seems pretty safe. I expect
> to be around for a while.

If they are willing to, it seems _a lot_ cleaner to create a project  
account than to use an individual's login for the data.  
(When we were processing the g-2 2000 data, it was stored in my 
account, and I had to remember to type my password at least 
every 24 hours into "kinit" to renew the access to the archive.  
If I left Internet range for more than that time, the production 
died.)

-- Fred

-- Fred Gray / Visiting Postdoctoral Researcher                 --
-- Department of Physics / University of California, Berkeley   --
-- fegray@berkeley.edu / phone 510-642-2438 / fax 510-642-9811  --