Super User is a question and answer site for computer enthusiasts and power users. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

We run some applications in our Red Hat 5 servers. An application running on client machines accesses on a mount on this server. This application is used by users via X on windows PCs. There are data of 2009, 2010 and 2011. But they are not processed and not used. We sometimes access old data for preparing reports etc. We add new data each month and work on this new data every month. For example in January 2012 new data will be loaded on server and it will be processed by an application. But mount size is 80 GB and it often gets full. And we delete some unnecessary files from previous months. But no unnecessary files are left. Now we have 2 options: Either increase the disk size, or move the old data to another disk. I couldn't decide which option I should choose. I would like to know which option will be better for the application's performance.

Old data is around 70 GB. Current disk capacity is 80 GB. Which is better in terms of performance and readability:

1- Move old data to another disk and so that there will be around 7 GB of data in current disk and 73 GB free disk space. 2- Increase capacity of current mount to 500 GB disk capacity and there will be ~490 GB free disk space.

On which option do you think our application may run faster?


share|improve this question
Neither option is going to heavily affect speed. I'd look at other things that may affect the overall system. Like having just one disk for everything makes backups rather easy. Having multiple disks just means you'll need yet another disk for backups. – surfasb Dec 23 '11 at 9:42
up vote 2 down vote accepted

Using a new disk will be faster because drive technology gets better over time - including caching and seek speeds.

But an increase in capacity brings its own performance increases too, simply because of higher data density.

Because for a given revolution of a disk, the heads on a higher capacity drive will pass over more data than a lower capacity drive, and so can transfer more data per revolution. There are many other factors at play here, but this one is significant enough to go with the second option.

share|improve this answer
You mean increasing disk capacity is better. But there will be more data in 2nd option. Won't it be difficult for the application to read data amongst alot of files? – alwbtc Dec 23 '11 at 8:38
@alwbtc linux filesystems don't tend to clump files together, but prefer to spread them around and leave gaps to avoid the need to defragment files. So the adding and removing of files doesn't particularly impact performance. – Paul Dec 23 '11 at 8:48
not with modern filesystems for the most part. Modern filesystems do a pretty good job of keeping files defragmented. – Lamar B Dec 23 '11 at 8:49
So we can go with the 2nd option right? If we choose 1st option, we will not use a "new" disk, we will move old data to another disk, and continue running application in current disk with 80 GB disk space. Could you please tell me the disadvantages of 1st option? – alwbtc Dec 23 '11 at 8:51
@alwbtc Your question was which of the two options would provide better performance, so the disadvantage of option 1 is that it would not provide better performance – Paul Dec 23 '11 at 9:51

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.