Exclusive: The Dual Actuator/Head Harddrive

[supere123]

And for that matter, why don't we change SATA to include multichannels. hmm.....Sata cable with 4 channels.....hmmm......4 SSD's in 5.25" enclosure with 1 multichannel SATA connector. Hm....Configured for Raids of your Flavor...hmmm....For that matter, why don't we abandon electron data transfer entirely and move to faster photonic transfer....hmmm....

 

[CerianK]

[citation][nom]CerianK[/nom]That explanation is irrelevant, since the data from each head would be buffered and merged independently for reads and split and synced independently for writes. Only the complexity of the controller would change... current hardware (actuators, heads and platters) would be compatible, as is.[/citation]
My apologies in that criticism. Each head would indeed need micropositioning ability to keep all heads aligned simultaneously. Instead of expensive independantly controlled acutator arms, perhaps nitinal wires similar to control horizontal motion, similar to the way somebody was implementing a similar feature to contol head gap to the platter several years ago.

 

[Guest]

What about adding multiple R/W heads on each arm, at a certain distance away from eachother?

 

[we_are_theBorg]

CerianK, You are a welcome breath of fresh air in a forum filled with dilettantes.

I've been wondering why drives with multiple platters don't use their multiple R/W heads simultaneously for "striped" performance improvements. It has always struck me that the ability to "multiplex" the encoded binary data would be a surmountable problem, but I had not taken in to consideration the realtime stroke by stroke adjustments that must be made in each heads tracking independently every time a track must be accessed. This too, however seems like a solvable problem. Access time might go up, and for random IOPs the drives might not perform well, but for sequential access those things would *scream*... Well, perhaps it's not worth it for the mainstream drive manufacturers to mess with technology so far outside of the conventional when the SSD is in the process of eating their lunch, but since high capacity (read, multiple platter drive) is the area where magnetic media will be competitive the longest, it makes sense for the companies wedded to magnetic technology to find novel ways to make the performance more competitive with SSD technology...

I'd be terribly, terribly curious to know what the folks at the big drive manufacturers think about the possibility of such an implementation...

 

[trevorjs]

It's about time to revisit this isn't it? Dual controller/head/actuator should be an easy concept and would actually reduce power consumption while still offering higher performance with a the same footprint. Today's technology can do it. If anything, make the platters themselves smaller to accommodate any additional space needed to fit in a 3.5" standard form. There really is no reason not do this except to NOT sell more of another product to achieve the same result which is just bad ethics.