SSD Reliability

[Rugged 0]

Are you not updating the thread anymore? I think if you did you would find the Samsung 470 is going to be right up there with the Intel. On NBR they did a poll as well and it turned up some very intersting information.

Thanks!

[Brian 157]

Samsung has a lot of advantages since all components are in house. It's not a top performer, but what's performance without reliability?

[dhanson865 0]

Are you not updating the thread anymore? I think if you did you would find the Samsung 470 is going to be right up there with the Intel. On NBR they did a poll as well and it turned up some very interesting information.

Thanks!

Not at this time. Might not bother again.

I spent a lot of time an effort on two sites updating threads like this but I wasn't getting paid to do it unless you count complaints and arguments as payment. No good deed goes unpunished.

From the current uproar about issues with 6 Gbps drives the Intel G2 drives are the simple safe choice right now but they are also nowhere near leading edge performance wise.

The newer drives all have issues right now so it'd be hard to recommend one to someone that isn't willing to deal with those sorts of issues themselves. Before the media twisted the word those people were called "hackers", now I guess "enthusiast" or "techie" is a safer term.

[robot 0]

Not at this time. Might not bother again.

I spent a lot of time an effort on two sites updating threads like this but I wasn't getting paid to do it unless you count complaints and arguments as payment. No good deed goes unpunished.

From the current uproar about issues with 6 Gbps drives the Intel G2 drives are the simple safe choice right now but they are also nowhere near leading edge performance wise.

The newer drives all have issues right now so it'd be hard to recommend one to someone that isn't willing to deal with those sorts of issues themselves. Before the media twisted the word those people were called "hackers", now I guess "enthusiast" or "techie" is a safer term.

Intel just released new firmware for 320 series, claiming that they fixed the 8MB bug... so maybe now this is also a relatively safe choice?

[Axl 0]

Since this discussion is alive at the moment, any opinions on Micron/Crucial SSD reliability? It's way too early in their lifespan to read much into reliability, but I've seen mostly 5 egg reviews so far. I've been using a 64 GB m4 for almost a month. It's noticeably quicker than my 9 month old OWC 40GB Mercury Extreme, which itself is a great system drive (doesn't write very fast after 9 months without AHCI; Win7 sees it as SCSI; older nForce chipset in that system.) Crucial has been getting the best out of Marvell controllers lately. And the SLC P300 looks like a beast - impressive video Micron has of a single P300 running IOMeter vs 15k SAS array. But what about reliability? Personally, I think I feel safer playing my cards with Crucial than with Intel or Samsung, but only time can tell. I certainly agree that making all components in-house could (and most likely will) lead to higher compatibility.....do we think this has the same effect on reliability?

[mike2h 1]

i am highly interested in your question(s) to. i am debating wether to use a samsung 470 64 for a gane drive but the m4 looks very nice. the samsung, from my research seems t ohave ver few negs about it, which is really good considering its age. the m4's potential reliabitlity is still an unknown. at least for me...

[Beenthere 0]

From the Aug. 2011 Best SSD review at Tom's Hardware:

"There's no debating whether SSDs offer blistering performance. That that doesn't really matter if you can't trust the device holding that important information. When you read about Corsair's Force 3 recall, OCZ's firmware updates to prevent BSODs, Crucial's link power management issues, and Intel's SSD 320 that loses capacity after a power failure, all within a two-month period, you have to acknowledge that we're dealing with a technology that's simply a lot newer (and consequently less mature) than mechanical storage."

[MRFS 20]

I'm watching the SSD marketplace very closely, and everything I read above still convinces me to stay with HDDs,

until the SSD world improves a LOT!

Now hear this ...

I'm thinking of writing an unsolicited proposal to Western Digital to consider a RAID Edition 5 ("RE5") line of their HDDs.

Here's why: I can't find a 6G WD HDD with 64MB cache that also supports their Time-Limited Error Recovery ("TLER")

which they highly recommend for all RAID arrays, unless I buy their expensive SAS HDDs, or spend a lot on Velociraptors.

For example, at the WD website they warn against using multiple Caviar Blacks in a RAID array, because

those HDDs don't support TLER. The other reason is that, generally, higher performance comes with higher cost

and with higher capacity too.

So, imagine a WD RE5 HDD with 64MB or 128MB cache, 6G interface, 7,200 rpm, and perpendicular magnetic recording,

but in relatively small capacities e.g. 64GB, 128GB and 250GB -- targeting the SOHO market sector (initially).

With the right pricing, 4 x HDD caches @ 64MB = 256MB in RAID 0 (HDD caches are effectively additive in RAID 0 mode)

The closest I can come to this goal is the WD1500HLHX:

http://www.newegg.com/Product/Product.aspx?Item=N82E16822136928

... BUT the cache is too small, and for that price ($120) the capacity is too large for what I am suggesting here.

For comparisons, we have measured 283 MB/second with 2 x WD5003ABYX in RAID 0, using the ATTO benchmark:

that HDD has a 64MB cache, 7,200 rpm, PMR, but only a 3G interface.

I would jump tomorrow for an inexpensive 64GB version with PMR, 7,200 rpm, and

either a 64MB or 128MB cache, as long as its interface were also the 6G standard

(600 MB/second).

This "RE5" series would be designed specifically for RAID arrays where TLER is recommended,

and where the combined HDD cache total creates a large effective hardware buffer that

helps to eliminate the need for a dedicated I/O buffer on-board expensive RAID controllers.

Now, let's look at some realistic prices: the WD2503ABYX retails for only $70 at Newegg:

http://www.newegg.com/Product/Product.aspx?Item=N82E16822136797

Let's reduce the capacity to 64GB, keep the cache at 64MB, keep TLER, and up the interface to 6G:

WD should be able to sell such a "RE5" HDD at an MSRP at or near $75 USD.

So, 2 of these in RAID 0 cost $150 and should exceed 290 MB/second sustained.

4 of these in RAID 0 cost $300 and should exceed 500 MB/second sustained.

Our experimentation here was able to reach 615 MB/second, but that speed

required 4 x Hitachi 15,000 rpm 2.5" SAS HDDs at $250 each ($1,000 just for HDDs

and not counting the cost of a Highpoint 2720 controller).

Summary: Until SSD technology matures a lot more, RAID arrays of relatively small

but relatively fast rotating HDDs can perform quite admirably for much less money

and with much greater reliability than current SSDs are able to demonstrate.

P.S. Just one man's opinion here: your responses are all welcome!

MRFS

[mike2h 1]

i may be wrong but i dont think any 7200 hd can saturate sata2 let alone sata3. have u considered short stroking re4's in raid 0? since i am going ssd, i am going to do that setup on my wifes comp. check this guy out about middle of page- http://www.newegg.com/Product/Product.aspx?Item=22-136-697&SortField=3&SummaryType=0&Pagesize=10&PurchaseMark=&SelectedRating=-1&VideoOnlyMark=False&VendorMark=&IsFeedbackTab=true&Keywords=%28keywords%29&Page=2 - done all the hard work already:)

[MRFS 20]

> i dont think any 7200 hd can saturate sata2 let alone sata3.

Yes, you are correct on this detail: the fastest HDDs are

hoving right around 150-160MB/second, which is roughly half

of the 3G headroom (300 MB/second).

However, some of a RAID 0's raw performance comes from

the combined size of the multiple HDD caches, and

the larger that combined cache total, the better the

average performance.

When files are already resident in that combined cache,

the interface speed starts to make a much larger difference,

particularly when upstream PCI-E bandwidth is 2GB or 4GB per second

(x8 PCI-E Gen2 @ 500MB/second).

What I'm proposing to Western Digital is a compromise on cost

and capacity, that offers the SOHO sector an opportunity

to enhance performance while maintaining reliability

without the apparent risks, and costs, that come with current SSDs.

In other words, give the SOHO sector a chance to approach

"enterprise" performance without needing to buy relatively

large capacity HDDs and without needing to pay the

relatively high prices of those large capacity HDDs.

Also, for very practical reasons, when one member of a

2 drive RAID 0 fails, yes all the data is lost, but that

is also true if a single HDD fails. Likewise, that one

failed member must be replaced, and the data restored --

"same difference" as we used to say in college:

the failed drive must be replaced and

the data must be restored, in both instances.

Particularly for large serial files, like drive images, videos, music,

the cumulative wear on each member of a 2-drive RAID 0 is 50%;

the cumulative wear on each member of a 4-drive RAID 0 is 25%

of the same wear on a single spindle.

People who argue that RAID 0 arrays are more likely to fail

are not addressing the reality of conditional probability

(a mathematical and statistical concept), nor have they

confronted the FACT that RAID 0 members each experience

less wear as compared to the same workload on a single spindle.

Conditional probability is the statistical likelihood that one member fails,

given that another member of the same RAID array fails: nobody has been

able to quantify that likelihood, yet many people repeat the same

theory as if it were a fact.

A RAID Edition 5 series in the 2.5" form factor could easily

support a 64MB HDD cache, PMR, 7,200 RPM, in relatively small

capacities, to allow the SOHO user to build high-performance

RAID 0 arrays that should cost much less than current SSDs,

all things considered.

And, 4-in-1 enclosures for 5.25" drive bays are starting

to proliferate e.g. Enhance Technology's x14, Thermaltake,

Icy Dock, and the "also rans" like Athena's defective unit.

Hitachi has succeeded in building reliable 2.5" SAS HDDs

with 64MB cache that spin at 15,000 rpm: we have 4 of them

running very reliably in one experimental workstation:

we use that RAID 0 array to read and write the contents

of a 10GB ramdisk, using RamDisk Plus from SuperSpeed LLC.

(That software saves ramdisk contents between shutdown and startup.)

So, I have no doubt that what I'm suggesting is technically feasible for WD:

the decision to do it, or not, is really a marketing decision,

more than anything else.

And, WDC recently purchased Hitachi's HDD division, so the technology

is already "in house".

MRFS

[ntxstormchaser 0]

What are the options for information recovery if an SSD crashes? I have heard that if a SSD crashes the data is lost.

[ntxstormchaser 0]

I am glad to see the reliability rates are coming up. I am getting more and more comfortable with the idea of an SSD in my mobiles.