Hi

We have a customer running 1TB diskpool. But, I cannot understand the 
reason for not going to tape. Going to tape is formost cheaper, but its 
also easier to manage a volume-based information store.

When setting up this large diskpool, you dont have to use a deviceclass of 
file. You can set it up as a normal storagepool with device=file. Or, 
which probably would be easier to manage, split it up into smaller 
storagepools, perhaps 1TB per diskpool on a 10TB disk. This will eliminate 
the need for reclamation and other sequential-based processes.

However, if your not doing remote copying, I would strongly recommend 
having a copy storagepool..........

Best Regards

Daniel Sparrman
-----------------------------------
Daniel Sparrman
Exist i Stockholm AB
Propellervägen 6B
183 62 HÄGERNÄS
Växel: 08 - 754 98 00
Mobil: 070 - 399 27 51




"Hokanson, Mark" <Mark.Hokanson AT WESTGROUP DOT COM>
Sent by: "ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU>
2003-01-09 16:15
Please respond to "ADSM: Dist Stor Manager"

 
        To:     ADSM-L AT VM.MARIST DOT EDU
        cc: 
        Subject:        TSM Disk Pool Management


We are considering creating an extremely large TSM disk storage pool (IE:
1-10TB). The goal being to eliminate going to tape. Does anyone have any
really BIG disk pools? Are there tools available for managing large TSM 
disk
pools over time. (IE: Reclamation, Aging Data, etc) The documentation from
Tivoli suggests that we need to set it up as a device type=FILE. What are
the drawbacks using this approach? Is performance dramatically impacted?
etc...

Thanks in advance,

Mark Hokanson
Thomson Legal & Regulatory