HDD to Morph Into 'HRD' to Beat SSD Comments

Author Gordon Kelly
Published 25th Jun 2009
HDD to Morph Into 'HRD' to Beat SSD

Comments for HDD to Morph Into 'HRD' to Beat SSD

« Read the Full News Story

comment Xiphias said on 25th June 2009

It's certainly interesting technology, even if it does have a stupid name. Fairly high power consumption too, the latest generation of SSDs are 1-2W.

They're claiming 160,000 IOPS for both read and write which is interesting, unfortunately they don't specify whether it's random or sequential. If it's both then that'll blow SSD's puny 500 I/Os for random writes right out of the water.

But still, Glass? Isn't that going to be fragile or heavy? I think notebooks may be sticking to SSDs even if this proves to be a better technology for desktops.

comment xbrumster said on 25th June 2009

too many fussy new names. can't catch up with all those new techno junky. Give me a super fast computer I'll be happy.

comment Ed said on 25th June 2009

Sorry, what's fussy about HRD? Along with HDD and SSD that makes for three terms that cover pretty much the entire mass storage market. Hardly difficult.

As for the glass comment, assuming the drive is cushioned there shouldn't be any problem. If you think of the mechanics of a HDD and how fragile they then realise they survive perfectly well in current laptops. This is just going to improve that further because there's no moving parts. That said, yes it will be a bit heavier.

comment Kingsley said on 25th June 2009

@Xiphias - glass platters are common place in HDDs now http://en.wikipedia.org/wiki/Hard_disk_platter so no worry there.

comment John McLean said on 25th June 2009

Presumably like SSDs power consumption at idle will be negligible? I also imagine it is virtually silent, and should be more responsive than HDD as there is no need to spin up from sleep/cold and no need to move heads large distances (i.e. super-low seek times). Ability to read from many heads in parallel also a boon, but will need some clever controller circuitry as the whole data substrate moves as one, so each head must be in the same relative portion of its "zone" at a given time. I imagine long term no reason not to have multiple platters which, unlike conventional HDDs, could be addressed independently (i.e. they needn't move synchronously).

comment Hallainzil said on 25th June 2009

For me, provided I've understood it correctly, is the potential for growth here. I would think that the growth in speed could be exponential too, as it seems they would merely have to allow more read/write heads to be active at once.

If it offers this sort of performance with 64, imagine 128, 256 or even 1024. Very clever idea.

comment smc8788 said on 25th June 2009

@ Kingsley - Another example of the dangers of Wikipedia. Although some glass may be used in platters, the substrate (which is what makes up the vast majority of the platter, and is what the media layer is printed on) is generally made up of a glass-ceramic composite.

However, I'm not sure there would be a substantial difference in weight, as a quick Googling reveals that glass has a specific gravity of around 2.5-3 depending on type, while most ceramics are around 3-4 (again, depending on type).

comment Kingsley said on 25th June 2009

@smc8788 - Hmmm. The link was intended as a general indication that "glass platters" is not a risky, new technological solution as suggested by Xiphias. If you want another link, try http://www.storagereview.com/guide2000/ref/hdd/op/mediaMaterials.html - I think "glass" is used as a sort of catch-all name in this case, rather like "LCD" in flat-panels covers a multitude of different technologies which differ at some level of detail.

Add Your Comment

add comment Add your comment

You must be logged in to comment. Login or register here.