Seagate: "Hybrid Drives Will Outlive Solid-State Drives"

Where does this myth come from? Secure Erase has only one purpose and that is to completely sanitize a drive so that no sensitive data can be recovered from it. It exists for data security purposes and is a single pass overwrite of every sector/block of the storage disk (including bad sectors/blocks) with random junk data. The reason why it is 'secure' is because it is a mechanism built into the drive itself and it is executed at the drive level, not by software on the drive meaning that it is much more difficult for malicious software to screw with the process.

I don't understand how secure erasing is supposed to improve SSD performance since the degradation issues you talk about are to do with write amplification?

 

Oddly enough I have a drive with TRIM enabled, and I've still seen a performance decrease since I bought the drive, TRIM helps but its obviously not the be all and end all.

As to whether a SE or similar actually makes any difference?
Who knows, but I'm pretty sure I've seen it mentioned over on the overclockers and OCZ forums too, not to mention I've spoken to other users who have specifically stated that even WITH an exact carbon copy image of the drive reimaged back afterwards, they have experienced performance increases AFTER running an erase.

Whether this is MEANT to be the case or not, for some users it obviously works, or it wouldnt be recommended so much. Obviously its quite possible that how valid this tactic is varies between drive controllers etc.
Unfortunately SSD does seem to be a bit of a mythbusters paradise right now, there's a lot of 'do this/do that to achieve best performance' out there.

 

I hope to see Samsung and Hitachi implement this technology. In my opinion the best HDD manufacturers.

 

Seagate probably has a patent on their hybrid methodology, though I wish we see HDD manufacturers just implementing a full SSD on one side of their current HDD drives similar to ODD manufacturers.

 

Not quite true. I have witnessed it first hand on several occasions. Secure erase helps so that all the blocks are cleared out, and depending on the SSD this can be a significant improvement. With a Samsung SSD I had, performance became completely dismal, I mean 7200RPM HDD performance. I did a secure erase and it was back to near fresh performance. Same thing with my Intel 80GB. I did it once, and am about to do it again. Performance has dropped significantly. TRIM I guess helps, but a drive can still significantly degrade over time (and limited time too). Probably best to use as large an SSD so you can delay the inevitable longer.

They probably do have a patent, but patents are fickle things. I'm sure it specifies their algorithm, and not just a generic "hybrid" SSD, so any other manufacturer who uses a slightly different method wouldn't violate that patent. I don't know that something along the lines of "Integral use of NAND as a smart cache to improve hard drive
performance" would fly as a patent. It would have to detail how it improves it and what hardware and software is used.

However, it is curious why WD or Toshiba or other HDD manufacturer hasn't come forward with their version. It clearly is a marked improvement which laptop drives could benefit from.

One side note, is I did not realize WD actually makes a laptop velociraptor 10,000RPM drive. If anything one would think WD would make use of a similar tech to gain significant performance improvements especially with the desktop velociraptors.

 

I bought my Nova 128gb in March or April, it was $300, now it was on sale this last weekend for $199. Price is already dropping. It will continue. Also, someone wrote earlier that they had replaced hdd's 5 times and none for malfunction, but all because of size/speed. I have done the same, I have not had a hdd fail on me, but I have changed out drives for performance and size many times. I use the old drive for backups. I doubt my ssd will outlive my need to replace it in a year or 2 for bigger, better, faster. Just as I did with hdd's. Also, several had mentioned that most lappys have one drive, but more and more people are using their laps to replace desktops, nd more and more have dual (or even 3) drives. I have a ssd and hdd in mine, and more lappys are being configured that way. As the price drops, it will be more probable that more people will use both. The difference in speed is noticable.

 

I have 5 ssd's, two crapped out on me in less than two months of use. I now realize durability and reliability are two completely different categories. SSD's are durable, but so far not so reliable.

Also, the trend is for laptops to become smaller and lighter. Utilizing two drives does not fall into that trend. For a majority of users, web surfing, editing family photos and videos, and using Office documents are their primary function of a PC. Not many people will get a big and bulky DTR for that.

 

I apologize if I am wrong, however I follow the forum of others with my lappy, and there are sure a lot. Also, many of my friends and network acquaintances have gone mobile on the dtr. And many have gone to a dual drive/dual bay lappy. I get that netbook types have opened a large niche, but there are many more options for multiple bay laptops then ever before. I would say that in the older (older than 25-30 group) there are many more users of multiple bay laptops than maybe you would think. And they are the ones that I think will drive the ssd in the future. For us that are older, and need more than low end thin note/net books, we need the ability to be able to have all our computing needs available wherever we are, and sometimes that excludes small, underpowered books. Just because many of us are not on forums does not mean we do not have our needs, and as someone earlier posted, there is a lot of clout with us. Time is money to us, and saving time means making money. My favorite saying is "if it doesn't make dollars, it doesn't make sense". There are a lot of 30, 40 and 50 year olds well versed on computing (since it really became mainstream with us) that have higher needs than web browsing or email. Now do not think I am flaming you, I just am trying to let you see that there is a very significant group of older players that have these needs, need for more performance and power, speed and size, and work takes so much time for many that we do not always post like the younger ones. But our dollar buys a lot, and when I was trying to decide which dtr I wanted, there were very many choices. All of mine were between dual bay laptops. And I had quite a few to pick from. Now there are more, just in the last 8 months or so. But even in the netbook generation, many want speed. As for your 5 ssd's, I don't know what you picked, but I have had 2. I had a Kingson for about 5 hours, returned it because it just didn't give me what I wanted, then got a Nova 128, and have loved it. I have timed some tests in my work, and it is giving me some excellent numbers. And I am talking about time savings in my real world work, not benchmarks.

I do not think ssd's are for everyone, and I don't think everyone should go out and buy one. But for me, any time savings is big, and I am getting close to 33% time savings on my most time involved work, that is huge to me. Over the course of a year, my ssd will repay me several times over. That is where they will make, and keep making inroads over hdd's. Sorry for the tl,dr post, and I certainly am not being disrespectful or trollish in this post. I have a smile on my face as I post this, and I have read many of your posts wingnut, and think you have a very good grasp on technology. I just think you do not see a large silent group of users on this one. I have much respect for everything I have read from you. Disclaimer: all of my info on this comes solely from my observations, both shopping, talking and seeing. It is completely unscientific, and is not meant to portray anything other than what I see and hear. For all I know, I could know the only group of people that think like me, which is pretty much how we all are, we hang with those of like mind. Please do not try this at home. LOL.

 

Maybe there is no money in it?

They don't. That drive is not laptop compatible. Western Digital's Raptors have always just been enterprise 10k rpm drives with consumer SATA connectors. Western Digital's 10k rpm enterprise drives moved to the 2.5" form factor, so guess what, the Raptors did too. WD actually merged their 10k rpm enterprise and consumer drives with the Velociraptor, which is probably the only thing keeping the line around at this point.

 

I'm one of these elder folks here. But DTR's are still a niche market compared with the total pool of laptops. I don't have any links atm, but somewhere along the lines of less than 10%. Everything else is 15" or less with single drives.

 

That's far from statistically sound, as you probably understand. I owned about 7 SSDs, I've never had one fail. That doesn't mean anything either.

The only research I read about in this area said that for notebooks SSDs are more reliable than HDDs.

 

I know, but with over 2million hours MTBF, having two fail from different companies, seems like winning the lottery. But in a bad way. All I am saying is that statistically you can call shenanigans on the high reliability of claims if you have failures like that, where if you don't have any, it means nothing.

I'm saying from personal experience (been building and dealing with PC's for at least 30 years), in the last 10 years have had one hard drive fail, and it was several years old. You can understand how it can leave a bad taste in one's mouth to have two fail, new, in a short period of time. If you check out these forums as well as OCZ there's lots of others with dead drives. Something is obviously wrong. Plus I am not sure how they can really determine reliability considering how the tech is new and constantly evolving. Hard drives have a long history to draw data from.

 

There actually IS a place you can get some data from.

Go look at the Newegg ratings for storage products. In particular, note the percentage of 1 or 2 egg ratings, those are usually due to failures, if you look them up. The percentage of 1 or 2 egg ratings are very high for all SSDs compared to HDDs, except the Intel SSDs. Granted, it is biased due to the fact that people are more likely to give negative feedback, it still gives a sort of comparison between SSD and HDD reliability.

 

I guess it is bad luck or Intel is more reliable than competitors.

Intel put SSDs in their employees laptops worldwide and tracked them for a year (thousands of laptops).

Failure rates came out as follows (using previous study on HDDs as comparison):
HDDs: 4.47%
SSDs: 0.51%

http://download.intel.com/it/pdf/En...t_of_Notebook_PCs_with_Solid-State_Drives.pdf

 

Hmm, Intel who makes SSDs but not HDDs has a very positive report on Intel SSD's. Nooo. Say it isn't so! /sarcasm off

 

10 times more reliable by their data. If it's accurate I don't know but I bet SSDs are more reliable in notebooks than HDDs, for the simple reason they are more shock and fall proof

 

Right. But as I mentioned before there is a difference between reliable and durable. SSD's are more durable, reliability is still in question. Any manufacturer that releases "studies" is nothing more than marketing hoo haa. We don't know what brand of hard drives, what environments they were subjected to. Were the hard drives in the harsh vibration environments or in laptops more prone to being dropped? I mean for all we know they could have put SSD's in most of the corporate users machines which sit on the desk most days and hard drives in the field tech's machines. SSD's in machines with how many writes/day? Too many variables to take their numbers at face value.

If they want to argue durability, I'm all there. But for reliability, not sure I'm convinced. A failed drive due to durability concerns is different than one that fails on its own with little to no outside influence.

 

I think it's good to be critical of Intel's study.

Personally I would be even more critical of a study that only involved 5 samples.

 

Mock me all you want. I wasn't doing a study, that's the problem. I'm an end consumer spending top dollar for a product claimed to be more reliable than hard drives. Considering my experience or at least long term exposure to using hard drives, I have a right to express my dissatisfaction when two fail from the get go. And it shouldn't be discounted on account that I "only" have a "sample" of 5. How many do I have to own for me to say anything? 100, 1000, 10000? Sorry, but if you want to buy my SSD's for me, then I'll be quiet about it. Until then, I have a right to express my opinion.

Ignoring issues doesn't help anything. Addressing them does. I guess we should just ignore all users who have an issue with their drive because it isn't "statistically sound". I originally stated my dissatisfaction and the *fact* that I had two die on me. You continued to state that it isn't "statistically sound". When your "reliable" Toyota Camry dies in the middle of nowhere stranding you, do you give a crap about how reliable statistically they are? No you don't. It failed you, period. Confidence lost.

 

I wasn't mocking you. I did point out the inaccuracy, in my opinion, in your conclusion that SSDs aren't reliable. I understand you don't like that but I have the right to voice my opinion too.

Not really. I would also consider that Toyota Camries could be really reliable but that I may have had bad luck. I'd probably try to find some data to make a better judgement.

 

Mean time before failure ratings are based on the collective running time of very large groups. So if you have 2 million drives with a 2 million hour MTBF, your first failure will be on average one hour into operation. But another thing to say about these manufacturer MTBF ratings is that they are fairly hokey to begin with and should be taken with a very, very large grain of salt as they are completely unrealistic and are far from a sound gauge of a product's reliability for an end user.

 

Right but you're trying to discount the fact that I had two failures out of five drives I had and is irrelevant. I was stating fact, and you were somehow trying convolute it or something so that it wasn't true.

I know about MTBF, I had all that crap in my six sigma training, and dealt with it first hand with all the automotive repair warranties I had to manage. Either way regardless of MTBF, claims of SSD's as more reliable than HDD's so far are unfounded. I'm not looking for an argument. Only facts to back up that claim, and none from manufacturers with results that support their own products.

Even then, I initially wasn't looking for any data, just stating that I had two dead drives out of five. But it somehow turned into an argument of a fact. I dunno.

 

When I said 'statistically not sound' I meant that drawing general conclusions based on 5 samples is statistically not valid. A much larger set would be necessary to draw valid conclusions.

I have no doubt this happened to you and that you are telling the truth.

By the way, reliability figures for OCZ might be different than for Intel drives. I have no data in this area.

 

They use different controllers, possibly different NAND chips. Did you managed to RMA your dead SSDs?

 

In process at the moment.

 

Perhaps you had bad luck. I've never had a traditional hard drive fail in it's own mostly lightning strikes, power surges. I only own 2 SSDs so I couldn't tell you the reliability, but I can assure you almost all my systems in the future will have an SSD as a boot drive, whether it be 30-60 GB, it's a price I can live with.