• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

disk access question

D

draggoon01

Senior member
so sometimes when i'm doing something hdd intensive with one drive is makes access to my other physical hard drive really slow. same sometimes with optical drive as well. why is this? is it because the ide transfer is maxed out? will serial ata fix that so you can access two drives independently without slow down?
 
When you have two drives on one IDE cable, only one drive can be active at a time. So the data transfers have to switch back and forth between drives. If you're doing a heavy data transfer on one drive, it may end up using a larger proportion of the access time and the operating system's attention.

If they're on different cables, then it's probably just the OS having too much attention taken by the big transfer. The type of IDE controller and driver used can make a big difference. The CPU may be heavily involved in the transfer, which UltraATA is supposed to reduce but some drivers and controller types don't do it so well. (For example the nforce2 IDE controller, some people see 35% and higher CPU usage when using hard drives.)

SerialATA only allows one drive per port, so there shouldn't be any issue with one drive hogging the bandwidth. However bad drivers or controller design could still result in heavy CPU usage.
 
Ummm... doesn't bus mastering allow more than one device to use the IDE bus at the same time?... in which case you'd have to have a hard drive that can sustain at least 66 MB/s transfer rate to saturate an ATA66 bus... which isn't going to happen, even with a CD-ROM being accessed at 52X at the same time.

I could be mistaken, but I thought that's what bus mastering did.
 
Bus mastering allows a device to take control of the bus and theoretically allow it to transfer to another device on the bus without the CPU being involved. This applies to the PCI bus though, not to the channels on an IDE controller. IDE continues to be limited to only one device active on each cable, although they did eventually implement the ability for commands to be signalled to both drives no matter which one is actively transferring data.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top