General discussion


Breaking an NTFS Volume Set

By tjapkesg ·
At some point, the harddisk space on two of our servers was expanded. From what I've been told, this was done by a consultant, who simply installed 3 new drives in each server, and expanded two existing volumes with volume sets.

Both systems areconfigured the same as follows:

Disk 0:
| C: 1GB | 500MB | E:2GB | F: 5GB |

Disk 1:
| E: 8GB | F: 9GB |

Disk 1 is obviously the new array installed by the consultant. The volume set spanning across to the old array almost certainly has a performance impact of some level. One of the servers is running SQL server, and it has already been determined that drive access is at peak levels consistantly throughout the day.

Our current plan is to of course perform a full backup of each system and repartition the drives appropriately, and then restore the data.

I'm looking for any recommendations or precautions we should take during this procedure, or for any alternative methods of breaking up a volume set like the ones we have.

One approach I have consider is to do the backup, delete one set, partition space, move data off the other set to that space, delete the other set and again partition space for that volume. The data that was moved doesn't necessarily have to be moved back of course, but the data from the volume set that was originally deleted would be restored to which ever space was left.

Any thoughts or suggestions?


This conversation is currently closed to new comments.

1 total post (Page 1 of 1)  
| Thread display: Collapse - | Expand +

All Comments

Collapse -

you've got the right idea...

by ARG CIO In reply to Breaking an NTFS Volume S ...


The thing you have to look at, is why is drive access peaking? If SQL is on one of the servers, and Exchange 5.5, SMS, and any other "resource intensive" application is on that server, you may want to move it. SQL, and home directories(file serving) can be on the same server, but any other applications, and your gonna slow down. Make sure you span your applications, and try to spread the load. It may be that your SQL server is so heavily used, that your best answer is to upgrade to SCSI-160. It is not a cheap option, but having 100 users waiting 20 sec. 50 times a day, can get expensive too. If the SQL server is on DISK1 (which I assume is on at least three physical drives), then you may want to limit it to only the three new drives(maybe four), and span disk 0 across other physical drives. You may have too many volumes spanning too many drives, and causing unneeded reads. I hope this makes sense to you, good luck, and let me know how it turned out!

Back to IT Employment Forum
1 total post (Page 1 of 1)  

Related Discussions

Related Forums