Ivor O’Connor

January 17, 2009

A Disk Management Process

Filed under: Uncategorized — ioconnor @ 8:01 am


I’ve noticed boot times are highly dependent on fast hard drives. Or fast access to the data via good file systems as the above article proves. However I don’ t see any mention of predictive disk reads as I do for CPUs. No pipes. No threads. Not much of anything actually. Why is it boot data is not organized into contiguous sections that can be read via the equivalent of a single swipe? Instead we wait as each process requests its own data. We listen to the disk continuously move about. We wait and wait while the disk churns. Do we really have to wait another five years for a disk management process? One that organizes the data based on when we are likely to need it. One that organizes the boot up data so everything needed in the next few minutes can be read within the first split second? Is this really that complicated?!

No. It’s not. It’s not complicated at all. Every piece of information needed at boot time can easily be identified by any number of means. It can be organized in nice contiguous chunks. It can be read from disk with a single split second rotation of the disk.

A disk manager process along these lines is sorely needed. In the past when RAM and CPU power were limited this concept was not possible. Now however it’s an idea waiting to make somebody rich.


Leave a Comment »

No comments yet.

RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Blog at WordPress.com.

%d bloggers like this: