If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|
Thread Tools | Display Modes |
#11
|
|||
|
|||
"Why I Will Never Buy a Hard Drive Again"
Remapping (redirection from old bad sectors/blocks to the spare
sectors/blocks in the reserve space) is done in the firmware of the drive (although the OS can direct the firmware to flag a sector/block as bad to get the drive to do the remap). Before a reserved sector can get used, it has to be tested and, if okay, then the data is copied there and the old sector/block gets flagged by the drive's firmware. Even if you reformat the drive, the table on the drive will still see the bad blocks the firmware flagged as bad. Sectors/blocks that get remapped will NEVER be reusable by the OS or any process. The flagged blocks are marked bad for life. The drives already come with bad sectors or memory blocks. Those get masked at the time of manufacture when the devices are tested before released. A table of bad spots gets recorded in the firmware as a table in ROM on the drive's PCB. EEPROM is on the drive's PCB so more blocks can get flagged during use of the drive; i.e., the OS or tools can determine a block is flaky and get the drive's firmware to remap that block. That's why moving the PCB from one drive to another despite them being the exact same manufacture, product, and version will result in mismatched after-manufacture flagged blocks. What was good on one drive might use a bad block to where the PCB get moved, and what was bad on the source drive might have a block marked bad that was good on the drive to which the PCB got moved. https://www.mjm.co.uk/articles/bad-s...remapping.html Note the difference between primary remapping (p-list) at the time of manufacture and the G-list remapping performed during use of the drive. SSDs are a little different. Every write is a remap. That's part of the wear leveling algorithm to spread the writes out across the entire capacity of the SSD. You write some data in one place, read it, and when you write it then it goes somewhere else on the SSD. When a block is determined as bad, it get flagged as unusable and remains unusable thereafter for the life of the SSD. So, looks like I mixed HDD and SSD together regarding remapping. HDDs remap only when a sector is determined as bad and that causes redirection from that sector to the reserve sector. SSDs remap on every write as part of wear leveling, and blocks get flagged unusable forever thereafter if found bad. I've seen benchmarks where SSDs get slower on writes as more blocks are flagged as unusable, so I'm not sure why since remapping is always happening to move the data to somewhere else on every write. Users complain there is gradual degradation of performance of SSDs, and they are using TRIM and GC. Always remapping and flagging blocks as bad which can never be reused would seem to merely effect a reduction the capacity of the SSD during use as blocks eventually go bad, yet I've seen them catastrophically fail and it has been an artifact attributed to SSDs for quite a while. With blocks getting marked as unusable ever after which would reduce the capacity of the SSD, how is it the partition doesn't shrink? SSDs use overprovisioning: they have about 7%, or more, extra space than the OS can get at. When blocks get flagged as unusable thereafter, apparently a block comes out of the failed block reserve in the overprovisioning. So this is akin to how HDDs have reserve space to which bad sectors get remapped (in the drive's firmware, not in the file system). Seems the catastrophic failure of SSDs that others have noted is when the total of flagged unusable blocks exceeds the space in the overprovisioning for failed blocks. Any remapping thereafter would corrupt the file system within the partition. Health monitors can use SMART for HDDs or SSDs to determine if they are running out of reserve space. https://www.seagate.com/tech-insight...its-master-ti/ https://www.kingston.com/us/ssd/overprovisioning Where could a sector reside for a cluster within the file system if no more bad blocks can be flagged unusable? I have never attempt to adjust the overprovisioning of an SSD. If I up the overprovisioning, the SSD's capacity decreases which means [aggregate] size of my partition[s] would have to decrease. https://www.youtube.com/watch?v=bHf6rCDUTYU I have a Samsung SSD and also have their Magician software; however, I don't touch overprovisioning. It shows me my SSD has 10% for overprovision space. While the Youtube video shows how to remove overprovisioning, that looks stupid and dangerous. The articles that I've read about *increasing* the SSD's performance is to *increase* the overprovisioning. Also, with an increase in overprovisioning, maybe you get a larger reserve space, so more blocks can go bad before the SSD fails catastrophically (hopefully by becoming a read-only device but that means the OS won't load since it also needs to write). http://www.atpinc.com/Memory-insider...-wear-leveling Upon further reading, the progressive slowdown of SSDs that other experience (but not me) might be due to them piling too much into their SSD. At a certain point, the SSD will slow down as it fills up. https://www.howtogeek.com/165542/why...-fill-them-up/ https://pureinfotech.com/why-solid-s...ce-slows-down/ That explains why SSDs slow down as you fill them up. I keep my SSDs at low consumption; for example, on my home PC, I'm currently using only 20% of the SSD's rated capacity. It could be that the progressive slowdown noticed by others on their PCs with SSDs is due to them overfilling their SSDs, like beyond 70%. I've never really kept track of how full are the SSDs of other users, only that they complain their SSD has slowed down and nowhere as speedy as when they first installed it. Note that catastrophic failure for an SSD does *not* mean the drive becomes unreadable. The SSD will [hopefully] switch into read-only mode, just no writing. Since the OS want to write to its own partition, that's likely a cause for "catastrophic failure": the OS cannot load. So you can probably still get your data off a failed SSD (unless cause for failure is due to non-media components) but as a data drive. SSDs wear out and stop being writable devices but should still be readable unless failure was not due to the media (memory). Well, an SSD degrading into a read-only drive is a best case scenario for a failed SSD. Seems that is how they are designed. HDDs die due to failure of its mechanicals, not because the media went bad and why you can still get data off its platters or even move the platters into another housing. Ain't silicon voodoo fun? |
#12
|
|||
|
|||
"Why I Will Never Buy a Hard Drive Again"
On Thu, 9 Aug 2018 00:52:41 -0400, Bill wrote:
I've heard that the reliability of SSDs far exceeds that of the mechanical hard drives (for, in fact, an obvious reason--no moving parts). The "trim" software for my Intel SSD even provides an indication of the drive's reliability (I'm not sure how well that works). I do regular backups too. I get a little confused on these new array memory schemes, the 3D stacking of 2- or 3-bit address advantages. Presumably MLC isn't quite as volatile a restructure of technological accountability -- apart from caching advantages, if and when employed and neither to exclude SLC. Generally and apart upper-end drives reflecting that price premium when exclusively employed. And then there's also the whole controller issue, perhaps more recognizable characteristically, enough so for indications of established baseline performance. Crucial and Samsung would seem most of all dominating, although I must say all my SSDs also are theirs. Leaving the residual of the indicative subject to TLC NAND, which is presently going through marketing loops and spins. There's now a new all-time low hitting, every day, a SSD market for TTC. And they can very aggressive in purporting unique merits of both memory and controller structures, hitherto unavailable from a constraints of technological understanding. As if almost a sideshow to the deluge and onus of any cost-to-performance, obvious from both the warranty intention, synonymous to total terabytes rewritten, a drive hypothetically is projected to withstand. A wide field, as it is now, covering not an inconsiderable amount of means available, once seen from a focal objective where industry is marketing an important distinction of SDDs now, and those SDDs which shortly have preceded them. And for these "everyday" sale TLC items, your nickel indeed will stretch far, squeezed, before the buffalo ultimately squats;- and, as is in keeping the plurality of things, there will be a dearth of realworld reviews, among those scant few, too conspicuous to not include, whom invariably express both displeasure at near or immediate failure, upon assuming receipt of their "everyday" TLC SDD, usually with a side-barb towards a faulty warranty mechanism skewed from industrial clout. With actually for a moment's pause, if hardly given to reassess, that neither Samsung, nor Crucial, would realistically ignore the same TLC 3D NAND technology, from a standpoint of applicable popularly they're favored when stepping out and onto the razor-edge's delineating this technological void. Less a matter of end- than deferring to the gloss-reviews from actual hardware sites, I find, where manufacturers traditionally supply subject merchandise for testing purposes, assessment and publicity. |
#13
|
|||
|
|||
"Why I Will Never Buy a Hard Drive Again"
Bill wrote:
mike wrote: How's the reliability? I'm still reading that they fail catastrophically without warning. I've heard that the reliability of SSDs far exceeds that of the mechanical hard drives (for, in fact, an obvious reason--no moving parts). The "trim" software for my Intel SSD even provides an indication of the drive's reliability (I'm not sure how well that works). I do regular backups too. You're the perfect customer for an SSD. You're mixing up reliability and wear life. Reliability consists of two components. Say a solder joint on the PCB fails. It causes the device to stop delivering the intended function. That's part of the reliability number. Let's pretend for the sake of argument, it's an MTBF of 2 million hours. In some cases, just the tiny power converter inside, making VCore for some chip, might dominate the reliability calc (you can't make a power converter better than about 10 million hours or 100 FITS). OK, well, what rate do bugs show up in the SSD firmware ? We don't know. We do know, that early SSDs "bricked" due to firmware. In some cases, the drive even "bricked" during a firmware update (but of course the owner backed up the data, making the situation not quite the same). In a system at work, our reliability expert (a guy with a PhD in the subject), warned that some large products we were selling, it was quite possible the software was dropping the system reliability by a factor of 10. Now the MTBF is down to 200,000 hours. You will find Seagate and WDC unwilling to factor this in. While our reliability expert argued for this, only field data could indicate how sucky our software was. ******* Wear life is different. Both hard drives and SSDs wear. In the case of the SSD, the mechanism is known and predictable. If you know the temperature when the writes were done, you know the temperature of the media over long-term life, you can make a reasonably accurate prediction of wear. (High temperatures anneal defects, but high temperatures might also shorten retention time.) Hard drives are different. The manufacturer won't admit to wear. The manufacturer won't prepare large quantities of drives, and simulate life conditions, and provide curves related to wear. But, third party studies have noted wear characteristics in the failure population curves. Instead of a traditional bathtub curve, drive failures have another shape in the graph. There are tremendous differences between various model numbers for this (things that might be noted by Newegg reviewers if a model is for sale for long enough). ******* Now, let's summarize: What do you have to know as an SSD owner. 1) Consider the history of the technology. You're doing basically what my PhD guy at work was doing, consulting a "field return data" log and noting brickage, brickage caused by bad firmware. For early SSD drives, you wouldn't touch them with a barge pole. Especially the ones with "predictable brickage", where the device fails after being powered for exactly 30 days. Owners who didn't hear about the 30 day brickage, might not have known (in time) that there was a firmware update for it, to be applied in advance. If it bricked and you had no backup (because it was "reliable"), well, "fool you once". Now you're learning. 2) Consider the wear life. The drives are taking fewer and fewer write cycles per flash location, as the technology "advances". The storage cells are getting "mushy". SLC, MLC, TLC, QLC. SLC is great stuff. Maybe 100,000 write cycles and 10 year retention. QLC might be 1,000 write cycles and ?? year retention. A Samsung TLC was showing signs of being "mushy", by requiring significant error correction inside (to the point it was slowing the read rate). Roughly 10% of the storage capacity on the drive, is reserved for ECC code storage, protecting the data from errors. That is a very high ratio, much highe than hard drives in the past. It's quite possible every sector has at least one error in it, corrected by the CPU inside before you get it. And now, they're just starting to ship QLC. 3) Consider the end of life policy. Not all the drive brands have the same policy. Some return an error on each write at end of life (as a cheap way of warning you), causing the SSD to enter "read-only state". That is a reasonable policy, helping to warn and cover people who refuse to make backups. Windows won't run on a read only device, so you'll be smothered in error dialogs. That will get your attention, and make you back up the drive. But Intel just "bricks" the drive, when the *computed* wear value is exceeded. With an Intel brand SSD, you had better be monitoring the "life remaining percentage" *very very carefully* . That's why the promotion in that Toms article above is particularly egregious. The dude is promoting an Intel QLC SSD (yuck!) which has a total-brickage end-of-life policy (double yuck!). What could go wrong ? If you're not paying attention, Beuler, you suddenly lose access to your data. Did you have backups ? No ? "Fool you twice". So, yeah, SSDs have no moving parts, and hay, they're "reliable". A stupid MIL spec calc prepared by the marketing department (not by engineers), says so. The firmware could have bugs. Not quantified in a MIL spec calc. They could have include field data in the MIL spec calc, but they'd be nuts to do so. No one is there to slap their fingers for failing to do this. The history of SSDs would mean dropping the MIL spec calc by a factor of ten. No marketing guy is going to allow that. But if your Sherman Tank is booted off an SSD, you can be damn sure two PhDs got into a spat about what the real reliability is. Between big companies doing business, the MTBF is "negotiated". The customer would say "hay, idiot, include firmware reliability in your calc". The wear life is tangible. There's an indicator in SMART. What is the brickage policy of your brand ? Pay attention! Is an SSD the same as a hard drive ? No, it is not. HTH, Paul |
#14
|
|||
|
|||
"Why I Will Never Buy a Hard Drive Again"
VanguardLH wrote:
Since I leave my computer running 24x7, I've *never* had the thermal creeping problem with connectors, memory modules, etc. The temperature variation on always-running equipment is not zero. Basically, any connector technology with a "walkout" problem, will eventually manifest. The DIMM slots have lock latches. The ATX main connector and ATX12V connector have latches. The newer SATA connectors have a metal jaw for security. You can have local heating effects, that have a higher amplitude variation per day, than the internal case air temperature. The Molex Aux connector on my video card, walked out on its own. And that's because the connector carries 5 amps+ when a game starts to play, and that caused the connector to heat up and walk out. When it got to the point that one pin was starting to separate (go ohmic), that's when the pin burned. It burned bad enough, to cause the video output to stop (the red "ATI warning box" appears on the screen, saying to plug in the cable). That's the first warning I got, that there was a problem. Since I didn't have the right connector in my junk box, I had to solder a pigtail to the video card (with a Molex on the end). That lasted until the card was retired. Even the solder balls on a badly designed video card can crack, just from heating from gaming. The fact you left the machine on at night, doesn't remove the variation when the card is used for gaming. This is why it's important that they select the correct underfill polymer to put under the BGA GPU package. While leaving a PC powered removes some reliability issues, it doesn't solve all of them. Paul |
#15
|
|||
|
|||
"Why I Will Never Buy a Hard Drive Again"
On 8/9/2018 9:37 AM, Lynn McGuire wrote:
"Why I Will Never Buy a Hard Drive Again" https://www.tomshardware.com/news/ch...ves,37563.html "It’s been years since I was willing to work on any PC that boots from a mechanical hard drive. Once you get used to the snappy response times and speedier gameload times of an SSD, going back to a hard drive feels like computing through a thick layer of molasses." But hard disks are still cheaper and more proven, despite of new encoding methods. You should copy important data to some old hard disks which use less risky encoding and could be more long-lasting than newer ones! It's also an interesting experiment to find out which old hard disk fail first. assumed that no "experts" sneaked into your home to do damage! -- @~@ Remain silent! Drink, Blink, Stretch! Live long and prosper!! / v \ Simplicity is Beauty! /( _ )\ May the Force and farces be with you! ^ ^ (x86_64 Ubuntu 9.10) Linux 2.6.39.3 不借貸! 不詐騙! 不*錢! 不援交! 不打交! 不打劫! 不自殺! 不求神! 請考慮綜援 (CSSA): http://www.swd.gov.hk/tc/index/site_...sub_addressesa |
#16
|
|||
|
|||
"Why I Will Never Buy a Hard Drive Again"
Rene Lamontagne wrote:
Do anyone here remember the Apple III walkout era? Rene Socketed chips ? Ouch. https://www.hardwaresecrets.com/inside-the-apple-iii/3/ Some fun for the owner I guess. I hope the cover comes off the unit easily :-) Paul |
#17
|
|||
|
|||
"Why I Will Never Buy a Hard Drive Again"
On Thu, 9 Aug 2018, Bill wrote:
mike wrote: How's the reliability? I'm still reading that they fail catastrophically without warning. I've heard that the reliability of SSDs far exceeds that of the mechanical hard drives (for, in fact, an obvious reason--no moving parts). The "trim" software for my Intel SSD even provides an indication of the drive's reliability (I'm not sure how well that works). I do regular backups too. But I've never had a hard drive problem. That goes back to 1993, when I go tmy first hard drive. I've moved on to different hard drives, but that's because of a different computer or wanting more space. But none have failed, not even the ones that had been used when I got them. I'm sure that when I get around to turning on that computer from 2003, which was used at the time, the hard drive will be fine. Though I splurged on a new hard drive, a 160g, about 2006. But it stayed on most of the time till I moved to a different computer in 2012. Hard drives became reliable at some point, and so cheap. And I'm yet to be convinced that an SSD is appreciably faster than a mechanical hard drive. Though it helps that I leave the computer on, so any "slowness" of booting is an iregular thing. Michael |
#18
|
|||
|
|||
"Why I Will Never Buy a Hard Drive Again"
VanguardLH wrote:
Ever look at the cost of a 2TB or 4TB SSD? Ouch!!! Note the phrase "boots from" in the Tom's Hardware clip. Run system and apps from a modestly sized SSD and store your big data, such as audio and video, on a big HDD. These days, it's the only way to fly. |
#19
|
|||
|
|||
"Why I Will Never Buy a Hard Drive Again"
Can I buy a 2 TB SSD for really cheap yet like a HDD price?
In alt.comp.hardware.pc-homebuilt Lynn McGuire wrote: "Why I Will Never Buy a Hard Drive Again" https://www.tomshardware.com/news/ch...ves,37563.html "It???s been years since I was willing to work on any PC that boots from a mechanical hard drive. Once you get used to the snappy response times and speedier gameload times of an SSD, going back to a hard drive feels like computing through a thick layer of molasses." Lynn -- Quote of the Week: "You feel the faint grit of ants beneath your shoes, but keep on walking because in this world you have to decide what you're willing to kill." --Tony Hoagland from "Candlelight" Note: A fixed width font (Courier, Monospace, etc.) is required to see this signature correctly. /\___/\Ant(Dude) @ http://antfarm.home.dhs.org / http://antfarm.ma.cx / /\ /\ \ Please nuke ANT if replying by e-mail privately. If credit- | |o o| | ing, then please kindly use Ant nickname and URL/link. \ _ / ( ) |
#20
|
|||
|
|||
"Why I Will Never Buy a Hard Drive Again"
On 08/09/2018 7:29 AM, Paul wrote:
VanguardLH wrote: Since I leave my computer running 24x7, I've *never* had the thermal creeping problem with connectors, memory modules, etc. The temperature variation on always-running equipment is not zero. Basically, any connector technology with a "walkout" problem, will eventually manifest. The DIMM slots have lock latches. The ATX main connector and ATX12V connector have latches. The newer SATA connectors have a metal jaw for security. You can have local heating effects, that have a higher amplitude variation per day, than the internal case air temperature. The Molex Aux connector on my video card, walked out on its own. And that's because the connector carries 5 amps+ when a game starts to play, and that caused the connector to heat up and walk out. When it got to the point that one pin was starting to separate (go ohmic), that's when the pin burned. It burned bad enough, to cause the video output to stop (the red "ATI warning box" appears on the screen, saying to plug in the cable). That's the first warning I got, that there was a problem. Since I didn't have the right connector in my junk box, I had to solder a pigtail to the video card (with a Molex on the end). That lasted until the card was retired. Even the solder balls on a badly designed video card can crack, just from heating from gaming. The fact you left the machine on at night, doesn't remove the variation when the card is used for gaming. This is why it's important that they select the correct underfill polymer to put under the BGA GPU package. While leaving a PC powered removes some reliability issues, it doesn't solve all of them. ** Paul Do anyone here remember the Apple III walkout era? Rene |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
"One Billion Drive Hours and Counting: Q1 2016 Hard Drive Stats" | Lynn McGuire[_2_] | Storage (alternative) | 7 | May 22nd 16 07:30 AM |
External USB hard drive showing wrong "Free Space" "Used Space" inthe Capacity | RayLopez99 | Homebuilt PC's | 3 | February 17th 14 08:40 PM |
USB bootable maker: Diff between "HP Drive Key Boot Utility" and "HP USB Disk Storage Format Tool"? | Jason Stacy | Storage (alternative) | 1 | April 21st 09 01:14 AM |
WinExplorer shows no "Used space/Free space" in properties for USB stick drive ? "Optimized for quick removal" error? | Joe deAngelo | Storage (alternative) | 0 | January 18th 08 01:28 PM |
Western Digital "My Book" - Replacing the hard drive | [email protected] | Storage (alternative) | 0 | July 8th 06 06:31 PM |