• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.
  • The forums have been upgraded with support for dark mode. By default it will follow the setting on your system/browser. You may override it by scrolling to the end of the page and clicking the gears icon.

Why SSD fail more than RAM ??

2018_newu

New Member
Joined
Nov 1, 2018
Messages
1 (0.00/day)
Both the RAMs and SSDs have no mechanical moving parts. So why my Intel SSD (5 years warranty) failed after 2+ years but my RAM works almost non-stop 10+ years without failing ??

I have a desktop for 10+ years which is ON 24/7 almost all this period with occasional restarts and OFFs. I use Windows 7 32 bit, 2 GB RAM. The SSD is installed in the same desktop for the last 2+ years.
 
Ram and flash (what SSD's use) memory are very different technologies. Flash has shorter longevity than ram, and is slower, but ram is volatile (looses memory on shutdown).
 
I'm not 100% certain, but if I had to guess, I would say the reason a solid-state drive wears out faster than ram is due to the need for a solid-state drive to retain its memory while powered off. If a solid-state drive were volatile memory like ram is my guess is it would last much longer do to strain inherent to the make up of the system which keeps memory intact while powered off in ssd drives.
 
Idk , but i've heard that formatting SSDs shortens the lifespan of the SSD.
 
I've never heard SSDs fail more often than RAM, at least modern SSD btw NAND (TLC?) wearing out is different.
 
I've never heard SSDs fail more often than RAM, at least modern SSD btw NAND (TLC?) wearing out is different.

It's the reason ram comes with lifetime warranties, and solid-state drives dont.
 
https://en.wikipedia.org/wiki/Flash_memory

Another limitation is that flash memory has a finite number of program – erase cycles (typically written as P/E cycles). Most commercially available flash products are guaranteed to withstand around 100,000 P/E cycles before the wear begins to deteriorate the integrity of the storage.[28] Micron Technology and Sun Microsystems announced an SLC NAND flash memory chip rated for 1,000,000 P/E cycles on 17 December 2008.[29]
The guaranteed cycle count may apply only to block zero (as is the case with TSOP NAND devices), or to all blocks (as in NOR). This effect is mitigated in some chip firmware or file system drivers by counting the writes and dynamically remapping blocks in order to spread write operations between sectors; this technique is called wear leveling. Another approach is to perform write verification and remapping to spare sectors in case of write failure, a technique called bad block management (BBM). For portable consumer devices, these wearout management techniques typically extend the life of the flash memory beyond the life of the device itself, and some data loss may be acceptable in these applications. For high reliability data storage, however, it is not advisable to use flash memory that would have to go through a large number of programming cycles. This limitation is meaningless for 'read-only' applications such as thin clients and routers, which are programmed only once or at most a few times during their lifetimes.
In December 2012, Taiwanese engineers from Macronix revealed their intention to announce at the 2012 IEEE International Electron Devices Meeting that they had figured out how to improve NAND flash storage read/write cycles from 10,000 to 100 million cycles using a "self-healing" process that used a flash chip with "onboard heaters that could anneal small groups of memory cells."[30] The built-in thermal annealing was to replace the usual erase cycle with a local high temperature process that not only erased the stored charge, but also repaired the electron-induced stress in the chip, giving write cycles of at least 100 million.[31] The result was to be a chip that could be erased and rewritten over and over, even when it should theoretically break down. As promising as Macronix’s breakthrough might have been for the mobile industry, however, there were no plans for a commercial product to be released any time in the near future
 
It's the reason ram comes with lifetime warranties, and solid-state drives dont.
Not every RAM comes with lifetime warranty, at least not here. RAM is also just as likely to fail as SSD, the reason for SSD failure though is generally different i.e. (SSD) controller or in rare cases firmware. I'll admit that SSD failure rates may be higher, but how much? In the meantime I have couple of crucial RAM sticks (8GB each) which don't work & have no warranty.
 
RAM is also just as likely to fail as SSD
incorrect.

SDD have components that manage & push electricity , and its a simple fact that they fail more frequently than RAM. if for no other reason than the fact that RAM is a simpler component than a SSD. I dont know the reason why they do for certain, but SSD's fail more frequently in every encounter, or instance i can recall it is the case.

unless my years of Exp, and personal encounters are totally out of the ordinary, and my read of the situation is totally out of whack...lets see what more informed members have to say.

EDIT*
i think its important to make clear, that when im speaking of RAM vs SSD's , when i say SSD's. i am speaking of the drive as a single unit, controllers, and all other included physical components (as opposed to Flash vs DDR), & in the context we are speaking, my guess would be the electrical components are the main culprit in regards to failure.

In regards to your crucial memory that is nonfunctional and unwarranted, I don't know what country you're in but that or where you bought it is likely the reason you don't have a warranty. Region specific coverage is nothing new, for me every stick of RAM I've ever purchased was guaranteed for its life time. However, a lifetime warranty on memory is kind of more of a Placation to consumers (like "all you can eat ,or unlimited refills, theres always fine print), than anything else. Due to advances in technology, and the sellers upgrading so regularly very rarely is the lifetime warranty on ram used which is also how I know it's very durable , I used to work with Dell as a technician, I have a fair bit of experience in component failure, at least in regards to Dell, HP and other major brands.

But im willing to admit im incorrect if i am, which i might be. Ive been out of the profession for some time, and my exp could be outdated.
 
Last edited:
incorrect.

SDD have components that manage & push electricity , and its a simple fact that they fail more frequently than RAM. if for no other reason than the fact that RAM is a simpler component than a SSD. I dont know the reason why they do for certain, but SSD's fail more frequently in every encounter, or instance i can recall it is the case.

unless my years of Exp, and personal encounters are totally out of the ordinary, and my read of the situation is totally out of whack...lets see what more informed members have to say.

EDIT*
i think its important to make clear, that when im speaking of RAM vs SSD's , when i say SSD's. i am speaking of the drive as a single unit, controllers, and all other included physical components (as opposed to Flash vs DDR), & in the context we are speaking, my guess would be the electrical components are the main culprit in regards to failure.

In regards to your crucial memory that is nonfunctional and unwarranted, I don't know what country you're in but that or where you bought it is likely the reason you don't have a warranty. Region specific coverage is nothing new, for me every stick of RAM I've ever purchased was guaranteed for its life time. However, a lifetime warranty on memory is kind of more of a Playcation than anything else. Due to advances in technology, and the sellers upgrading so regularly very rarely is the lifetime warranty on ram used which is also how I know it's very durable , not to mention I used to work with Dell as a technician and this is something I am fairly informed about at least in regards to Dell, HP and other major brands
Yeah & what the OP states is very likely a "component" issue, not NAND. The other arguments in this thread however deal with endurance, which as has been demonstrated, by that TR article among others, will literally take years before a decent SSD exhausts its P/E cycles.

However I stand by the argument that a modern SSD is just as (un)likely to fail as RAM, due to endurance, unless we have hard numbers that say otherwise.
 
Not every RAM comes with lifetime warranty, at least not here.
It ought to. Though it's not likely to be relevant.
RAM is also just as likely to fail as SSD...
Hell no. Not even remotely. Night and day difference. RAM lasts essentially forever. SSDs do not. Not now. Not ever.
...the reason for SSD failure though is generally different i.e. (SSD) controller or in rare cases firmware.
Right and wrong. Failure is always different(since they aren't even close to the same thing). Controller and firmware are not typically the things that "die" and/or cause and SSD to "die".
Macronix said:
Flash memory endurance is limited by the tunnel oxide degradation after repeated P/E stressing in strong electric field.
https://www.theregister.co.uk/2012/12/03/macronix_thermal_annealing_extends_life_of_flash_memory/

In layman's terms. The physical mechanism in non-volatile memory that allows the charge to be "trapped"(and thus the data to be stored when power to the device is switched off) is degradable. Volatile RAM has no such physical mechanism, and no such limitations. And is why it lasts virtually forever.

In the meantime I have couple of crucial RAM sticks (8GB each) which don't work & have no warranty.
Because:

a) They were defective from the factory.

b) You ran them outside of specs, or damaged then in some other manner.

c) For whatever reason Crucial's Limited Lifetime Warranty For Memory doesn't apply(or you're not aware of it).

d) All of the above.
 
Yeah & what the OP states is very likely a "component" issue, not NAND. The other arguments in this thread however deal with endurance, which as has been demonstrated, by that TR article among others, will literally take years before a decent SSD exhausts its P/E cycles.

However I stand by the argument that a modern SSD is just as (un)likely to fail as RAM, due to endurance, unless we have hard numbers that say otherwise.

You're welcome to your opinion, i respect that you disagree, I just happen to know your incorrect.
 
I can say this,, in 25 years of PC building .... I have probably had 5 bad sticks which I discovered on build day running memtest86+.

I, the past 8 years .... we have had 3 SSD failures, 0 SSD failures and 0 HD failures. The last one is misleading because we haven't put a HD in a build for 8 years. The last HD failure we had was an old SCI 10,000 rom drive which failed at about 8.5 years old.

One of the SSD failures was a warranty replacement failure of another one that failed. All were older models that would now be > 6 years old.

But my or anyone's personal experience is of little statistical value, and while industry data is readily available, unfortunately , it only reports failures between 6 and 12 months of operation.

Memory failure rates by brand for two consecutive 6 month periods:
  • Kingston 0,14% (0,45%)
  • Crucial 0,19% (0,36%)
  • G.Skill 0,54% (0,72%)
  • Corsair 0,90% (0,87%)
SSD failures by brand for same period
  • Samsung 0,20% (0,28%)
  • Intel 0,27% (0,34%)
  • Crucial 0,28% (0,42%)
  • Kingston 0,29% (0,26%)
  • Sandisk 0,62% (0,56%)
  • Corsair 1,67% (2,02%)
HD failures by brand for same period
  • Seagate 0,72% (contre 0,69%)
  • Toshiba 0,80% (contre 1,15%)
  • Western 1,04% (contre 1,03%)
  • HGST 1,13% (contre 0,60%)

The data must be viewed however in a particular context. most RAM failures will be discovered on Day 1.... RAM doesn't "wear out" . So for the RAM data, it can be assumed that actual failure rates were higher... with most happening oin the 0 - 6 month period and the data available reports only 6 - 12 months. For the SSDs, there are undoubtedly a fair number of DOA units, but most SSD failures in our experiences come about near the end of the warranty period.
 
I think I've only had one SSD fail. And never ram, unless if was bad when I received it.

And John I think you meant to use a decimal point and not a comma
 
And John I think you meant to use a decimal point and not a comma

That depends entirely on where you're from.

And I think I'm not wrong in saying DRAM is inherently simpler than flash storage (well, if silicone circuits can be called simple at all...) in that it is just made up of capacitors and transistors (one of each for one bit) in rows and columns. There's a reason it became so wildly used. SSDs need controller circuits as the cells don't have infinite reads/writes whereas DRAM has that, practically anyway if not technically.
 
It ought to. Though it's not likely to be relevant.
Hell no. Not even remotely. Night and day difference. RAM lasts essentially forever. SSDs do not. Not now. Not ever.
Right and wrong. Failure is always different(since they aren't even close to the same thing). Controller and firmware are not typically the things that "die" and/or cause and SSD to "die".

https://www.theregister.co.uk/2012/12/03/macronix_thermal_annealing_extends_life_of_flash_memory/

In layman's terms. The physical mechanism in non-volatile memory that allows the charge to be "trapped"(and thus the data to be stored when power to the device is switched off) is degradable. Volatile RAM has no such physical mechanism, and no such limitations. And is why it lasts virtually forever.

Because:

a) They were defective from the factory.

b) You ran them outside of specs, or damaged then in some other manner.

c) For whatever reason Crucial's Limited Lifetime Warranty For Memory doesn't apply(or you're not aware of it).

d) All of the above.
And that's why I asked for hard numbers & IMO views on this topic in the thread need to be taken with a pinch of salt. The reason being enthusiasts probably don't represent the typical failure rates across the industry.

Correct me if I'm wrong but what are the chances that overheating, due to OCing or bad airflow, mainboard or questionable power (supply) can damage main memory permanently? Perhaps over an extended period of time.

They were DDR3L SODIMM, ran in a laptop. One stick lasted about 12 months, the other a month or two longer. The laptop's doing fine but the sticks are not. The only mistake I did was buy them cheap off a reputed importer, whose website lists 1 year "seller" warranty & they were delivered seal packed.
 
Last edited:
I think I've only had one SSD fail. And never ram, unless if was bad when I received it.

And John I think you meant to use a decimal point and not a comma

It was a cut and paste right off the web site ... every 6 months the data is presented that way ... Either that's the way the Euros do it or its output from a csv file or some such
 
Back
Top