IBM Files Flexible Capacity SSD Patent

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Status
Not open for further replies.
[citation][nom]JohnnyLucky[/nom]Sounds like an interesting solution for some business enterprise situations. Would it also be practical for gamers and enthusiasts?[/citation]

Yes Very and for workstation use that is what I been looking for. I use a 30gb SSD for swapfile/virtual memory due to certain requirements such as video production that has a very large footprint that mechanical drives just can't keep up with. At the end of the day it is wow or some other games. All I need is high random 4k read & write and a long life. Mine seams to be going south after 3k hours of use.
 

PreferLinux

Distinguished
Dec 7, 2010
1,023
0
19,460
[citation][nom]sdflkj2lkjklfjskljf22[/nom]Man our patent system is so broken. This isnt a new idea, maybe with SSDs its new but not with HDDS. Weve been able to chose how to use the space forever. Choosing cluster size and partition size already changes how much real area you have on HDDs vs how much is reserved by the system, etc.I guess somehow using the words SSD and 'spare area' make this patentable. What a joke....[/citation]
You can still partition a SSD: this is completely different. The reserved space on a HDD isn't used for anything. The reserved space on a SSD is.

[citation][nom]mianmian[/nom]This method seems not as good as they claims. SSD cells can sustain 3000-5000 writes. Hiding some capacity would not help SSD life at all.[/citation]
They may sustain that many now, but get to a smaller size, and that number will decrease. Also this:
[citation][nom]hellwig[/nom]The problem here is that the drive (and the OS attached to it) thinks it has 128GB of usable storage space. If, say, 8GB have gone bad, it would really only have 120GB of storage space. The drive would need some way of telling the OS that, hey, I have 8GB of bad sectors, don't try to write more than 120GB of data. There's no such mechanism, and instead, when the OS writes data and there are only bad sectors left, the drive can't do anything and that data is lost and the drive corrupted. [/citation]
Compare it set to having 28 GB reserved, so the OS sees 100 GB: if you get 8 GB of bad sectors now, it is still fine. It won't actually extend the length of time you can use x GB for, it extends the length of time you can use the drive when it is full.
 

schmich

Distinguished
Sep 17, 2009
284
0
18,780
"reduce the size and reserve some of the memory cells as a safety net when other memory cells fail"
Isn't it possible to just shrink the drive size when cells go bad? Or mark that part of the drive as used? If not, does that mean when cells die on current SSDs the drive becomes unusable?
 
I think people are thinking this will impact capacity a lot more than it really will. Someone mentioned reducing a 128GB drive to a 64GB capacity. We're talking about a 64GB drive partitioning as 58GB instead of 51GB for the average user. That would work for most, but typically drives are over provisioned in case they're used in write heavy I/O applications.

That's a big deal because 51GB might not be enough for typical laptop useage, but that extra 10% you could add could make the difference. Same goes for 128GB drives.

I'd be much happier to buy a drive knowing that I could use an extra 10% free and they system would just give me some kind of warning if the over provisioned space was running out. Then you could just convert it to a dynamic drive and reduce the size.
 

kanaida

Distinguished
Mar 29, 2010
61
0
18,630
This feature has existed for 2 decades or longer.
Mechanical drives use a "low level" format, this type of format can modify the amound of reserved space. We just typically don't get options to play with.
It's also called "FIRMWARE". and guess what? there's updates for that too.
So while it's not something a consumer can do, it's definately "prior art/work". It's always been dynamic, the vendors just chose to not let us edit it directly because for the most part, consumers typically = retarded, except for the few that actually know *why and how* hardware works at all. It was just something you grew up with if you used a computer before the internet. It used to be common practice to know you DMA hardware address in hex, and IRQ (processor's interrupt request #), and the type of chip just to set up the sound to play wolfenstein lol...
 

alidan

Splendid
Aug 5, 2009
5,303
0
25,780
[citation][nom]bustapr[/nom]Seems logical and good. Too bad itll cost a fortune for while. See here Apple, this is what a REAL patent should look like.[/citation]
apple - lets patent the shape of the ssd, the weight, the way we installed it, and lets make damn sure to call how we connect it to something trade markable so we can charge per usage of the word.
ibm - lets patent the innovative way we found to increase the ssds life, so other don't reap the benefits of our hard work without compensation.

a clear difference
 

alidan

Splendid
Aug 5, 2009
5,303
0
25,780
[citation][nom]rosen380[/nom]Why would it be better to have, lets say, a 128GB drive that acts like a 64GB drive versus a 128 GB drive that acts like a 128 GB drive that degrades down to a 64GB drive at about the same time as the other one runs out of reserve memory?[/citation]

well lets say you set that 128gb drive to work like a 120 or 110gb drive. almost the same amount of space, and you have a safety net for when the drive runs down. a 64gb out of a 128gb would probably increase the drives life by 2x, but it also stands to reason, that in 5 years the drive would be replaced so why try to make it last 10.
 

kanaida

Distinguished
Mar 29, 2010
61
0
18,630
Oh and by the way, the technology is a great idea. I was just trying to get the point accross that the PSTO is nowhere near educated when it comes to technology. That's why patent trolls exist. I'm not saying IBM is one, but some patents shouldn't be awarded. If they didn't cost about $10,000 I'd be a patent troll too given that as far as computers go they can't figure out what 1+1 is
 

TNM

Distinguished
Jul 15, 2007
11
0
18,510
I haven't consumed the patent with all the details, but over provisioning has been normal in the SSD business for quite some time, this is often what separates server from consumer models.

Consumer drives often come with 7-10% spare areas while server models have 20-30%.

Intel has allowed consumers to customize over provisioning for quite some time and this was for instance described by them on the IDF in september 2009 (http://www.intel.com/idf/sf09/audio_sessions.htm MEMS003) and you can search for "intel ssd over provisioning" at G for whitepapers and technical info on how to over provision your intel ssd.

The patent is filed in October, so it would seem like IDF may provide prior art (but maybe there is something unique about the patent...).

We have consumer grade intel SSDs over provisioned in several of our servers. They definitely get reductions in performance under heavy load, but at worst case, they still perform much better than HDDs!

With the default provisioning they experience some very bad worst case scenarios however!
 

back_by_demand

Splendid
BANNED
Jul 16, 2009
4,821
0
22,780
[citation][nom]bustapr[/nom]Seems logical and good. Too bad itll cost a fortune for while. See here Apple, this is what a REAL patent should look like.[/citation]
Agreed, but the cost is not the issue, the issue is a real actual R&D project turning up a real actual patent.

In 10 years all SSDs will have this as default and the cost negligable.
 

back_by_demand

Splendid
BANNED
Jul 16, 2009
4,821
0
22,780
[citation][nom]rantoc[/nom]Shouldnt be much research put into this at all, firmware implementation however is another story...[/citation]
Sure, I guess you could have knocked this up over the weekend, right?

IBM do stuff like this all the time, they spend R&D money doing more that just reseaching, they actually develop as well. When they get a working version prototype they patent it.

A lot better than some generic theoretical paper idea that gets patented in advance.
 
G

Guest

Guest
this isn't even a new idea, been done for years, the idea of having reserve space that ill take over for bad cells/sectors....just shows how bad the patent system is in this country and why we have so many lawsuits over IP...change a name or two and you can get a new patent for an old idea, then sue everyone else that was already using it under a different name before
 

fir_ser

Distinguished
Apr 7, 2011
739
0
18,980
The good thing about these flexible SSD is the ability of the user to adjust the usable and reserved capacity, moreover predicting the life of the SSD drive is very important.
 

ProDigit10

Distinguished
Nov 19, 2010
585
1
18,980
it's probably best to make use of ALL cells from the start!
That's much better, than just saving some. In many cases an SSD would outlive a computer, and thus it's not necessary to reserve the much needed SSD space; space that perhaps in a year or two will be available for our budget!
 
Status
Not open for further replies.