EVA5000 disk space

[Karmen Lewis]

I have an EVA5000 running VCS2.006 with 84 76GB drives. If I configure all 84 drives as a single disk group, I lose over half the space. The max it wants to give me is roughly 2TB. It is configured for single failure, but I can't figure out why it is doing this. My other EVA5000 did not.

[Mike Naime]

The maximum you can create in any one LUN for presentation from the EVA is 2TB. Your OS limit may actually be lower than this size for what it can deal with.

I bet that if you created this 2TB LUN, and then try to create another LUN, you will once again be able to create about a 2TB LUN.

According to what you told us, you have 84 72GB (I'm fairly sure it's not 76's) drives in your EVA. This gives a rough total of 6048GB or 6 TB of disk space.

Now lets subtract about 1TB for 10% occupancy overhead, and sparing of 2 spindles. (Sparing is done in multiples of 2. So if your sparing level is 1, you are loosing 2 spindles.)

Now lets figure our redundancy level of the remaining 5TB. Assuming 5TB of usable space here, and a raid-5 redundancy level. We loose about 1/5 of the remaing space to redundancy. This takes us down to about 4TB.

Now, if we had created these at raid-1 level (Mirroring), we would have lost 1/2 of the space. Effectively giving us about 2.5TB of space.

What are your paranoia levels???
Sparing?
Raid Level?
Occupancy level?
Etc.

Note: I have a total of 168 146GB spindles in my EVA that is also configured as one disk group. (26TB raw space, with more on order) We get about 22TB (Before redundancy) of usable space out of it so far.

I am not making larger than a 1TB LUN for presentation to any system because the VMS OS will not deal with that large of a LUN.