Testbed Support for GridPP member institutes
> [mailto:[log in to unmask]] On Behalf Of Henry Nebrensky said:
> Also, what's a site - bear in mind an SE can fail without
> breaking a CE
> and vice versa. Does the availability apply to the CPU and to
> the disk separately, or to both together?
I would think it must be separately, or you could satisfy the criterion
as long as you kept an SE with one disk up 95% of the time, even if the
CE and the main disk server were nearly always down!
Stephen
|