An unfortunate tale nearly Samsung's SSD 840 read performance degradation

An barrage of reports emerged concluding September, when owners of the unremarkably speedy Samsung SSD 840 and SSD 840 EVO detected the drives were no longer performing every bit they used to.

The issue has to do with older blocks of data: reading old files is consistently slower than normal – equally deadening every bit ~30MB/s – whereas newly written files, like the ones used in benchmarks, perform as fast as new – around 500MB/s for the well regarded SSD 840 EVO. The reason no one had noticed (nosotros reviewed the drive dorsum in September 2022) is that information has to be several weeks erstwhile to show the problem. Samsung promptly admitted the upshot and proposed a fix.

Update (July 20, 2022): Samsung has released version 4.ix.7 of their SSD Magician software. Even though it's not mentioned on their website, this firmware update (DXT0AB0Q) is also available for the Samsung SSD 840, predecessor of the more popular SSD 840 EVO. We have published some follow up tests hither for your reference.

The First Firmware Update

Well-nigh a month later, on Oct 15th, Samsung released an updated firmware for the 840 EVO that covered both 2.5" and mSATA models (EXT0CB6Q and EXT42B6Q respectively). The update consisted of a two-stage process:

i) A new firmware with an updated algorithm for handling the inherent voltage drift that occurs in all NAND based storage devices as they historic period but is reinforced by how many bits the NAND stores:

  • In SLC NAND only one bit is stored per cell, this is nifty because information technology's very easy to read 1 fleck, information technology can either exist 0 or 1.
  • In MLC NAND 2 bits are stored per jail cell, so information technology gets harder to read but there's a cost advantage: you get twice equally much storage infinite from the same amount of NAND.
  • In TLC NAND 3 bits are stored per cell, and then once more the complication increases simply the advantage is that you can shop 50% more information vs MLC, further reducing costs.

Epitome courtesy of Anandtech

Co-ordinate to Samsung, the algorithm that adjusts the voltages used to read the NAND as information technology ages had a problem which meant that data previously written but never rewritten became harder and harder to read. The speeds to read such a file could plummet from 500MB/sec to below 50MB/sec, a 10x reduction in performance!

But this was a difficult problem to detect, because most benchmarking programs write new data that they then read dorsum, which circumvents the problem every bit it merely occurs on quondam data. However most information used past users is indeed onetime: your Windows installation folder, installed apps, your documents, game files, etc.

2) The 2nd stage of Samsung's new firmware with the updated algorithm mandated that all information on the disk should be rewritten to restore performance on older data. Since it took around eight weeks for the issue to get visible in the 840 EVO, this meant that nosotros could not fully know if Samsung's firmware worked or not until some weeks later.

A 2d Firmware Update: Reading Between The Lines

We couldn't know for sure if the firmware was a successful solution in the long term, and in fact the problem did come up dorsum. Samsung started to work on newer firmware (EXT0DB6Q), but this time with a different arroyo: instead of but adjusting the algorithm for reading old data, the disk would also continuously rewrite old data in the background.

It's not an elegant fix, and it's also a set up that will degrade the lifetime of the NAND since the total numbers of writes it'due south meant to withstand is limited. Just as nosotros take witnessed in Tech Report's extensive durability test there is a ton of headroom in how NAND is rated, so in my opinion this is non a problem. Heck, the Samsung 840 even outlasted ii MLC drives.

Equally of writing, the new firmware has just been released for the 2.5" model of the SSD 840 EVO, so users of the 840 EVO mSATA model still accept to be patient. It should also be noted that the new firmware does not seem to work well with the TRIM implementation in Linux, as this user shared how file organization corruption occurs if discard is enabled.

The route Samsung has taken with this latest ready is significant: the original trouble was not in the firmware of the drives, information technology'due south Samsung's TLC NAND which drifts in such a way that it's not possible to write a generalized algorithm that accounts for it. Thus by access we now know this is a fault inherent to the NAND used in the Samsung 840 EVO.

How About Other TLC SSDs?

Samsung claims the read performance degradation result simply exists in the NAND used past the pop SSD 840 EVO. All the same there are OEM versions of the drive that use the exact same NAND. For example, the Samsung SSD PM851 ordinarily seen on Dell products. Case in point, hither are users posting on support forums showing the exact same problem.

Then there'due south the "vanilla" SSD 840 which was the kickoff drive to use TLC NAND. Equally things stand up today no updated firmware has been released for this drive. Samsung Federal republic of germany admitted the problem exists on the 840, simply in Samsung'southward subsequent communications they accept always claimed that the effect does not exist on information technology. Hither's an excerpt from a recent Samsung Q&A posted at PC Per:

PC Per: Will there be a firmware update for the other Samsung TLC-based SSD models that accept also demonstrated this read performance issue? If so, which models and how before long will that firmware be made bachelor?

Samsung: This upshot had been reported for the 840 EVO SSD only.

Well, here is a Samsung SSD 840 "vanilla," or any you lot want to call information technology:

According to data nosotros've gathered from user forums:

  • The 840 EVO uses 19nm TLC NAND, it takes about 8 weeks to degrade.
  • The regular "840" pictured above uses 21nm TLC NAND, it takes nearly 40 weeks to degrade.

The test above was performed on a Lenovo ThinkPad T530 running Windows 7, using a Plextor M5M mSATA drive as primary storage and the Samsung SSD 840 connected as a secondary drive. A 64KB block size was used in Hd Tune, this limits the peak operation of the bulldoze to ~375MB/due south as you can meet where information technology flat lines. This apartment also corresponds to the free space area of 40GB on the bulldoze; it contains no information and thus is not affected by the degradation.

The first part of the disk has really poor performance, the reason is very simple: The drive was cloned from a regular HDD in a Compaq laptop, and the get-go partition contains a 13GB recovery sectionalization (really bad design for a mechanical HDD as that'southward where they perform all-time, but I digress).

That recovery partition has 2GB costless space, which corresponds with the lone fasten in functioning in the first 13GB batch of the test. Obviously the data in the recovery partition never changes and thus sees the worst functioning.

Delving deeper using the SSD Read Speed Tester tool adult past forum user Techie007 for the sole purpose of testing and visualizing the issue in the Samsung 840 drives, we get a graph of the performance of files in relation to how many weeks old they are. The graph includes all partitions because I mounted the recovery partitioning as a volume mountain point under the master segmentation:

Samsung 840 SSD Read Speed Examination Result

Looking at the graph, it becomes increasingly clear that the older the file is, as shown on the ten-centrality, the worse its read performance becomes.

Because I was testing with data that is several years old and the tester app express the visualization to 99 weeks, I took the raw data and put information technology on Excel to see what I came up with:

The x-centrality shows how old the data is in days, and the y-axis the read speed in MB/s. A healthy drive would have shown a flatter horizontal line hovering around the 500MB/s mark, only instead we get this mess!

For the sake of comparison, here'southward what a Samsung SSD 840 Pro looks like in SSD Read Speed Tester. Because the SSD 840 Pro uses MLC NAND, it doesn't suffer from the same degradation issues:

Samsung 840 SSD Pro Read Speed Test Issue: Steady Performance

Temperature Driving SSD Performance?

My criterion results saw wild fluctuations and I could not understand why. As it turns out the bulldoze is also sensitive to heat, and not in the way you might expect: the bulldoze really works better the hotter it is!

This is not then strange because temperature differences affect the voltage drift in the NAND. Simply now we begin to realize simply how difficult it must exist to go that algorithm correct… it's just a pity that it took Samsung just as long to realize as well.

The data for the graph to a higher place was gathered using SSD Read Speed tester when the drive reported that it was 40°C. The graph beneath loops the previous graph and the same test performed with the drive cooled to 15°C. You can run into how the worst-case operation is correct downward at ~50MB/s, a far cry from the optimal 500MB/due south for this bulldoze.

Information technology's entirely possible to presume that while the NAND in my drive performs better with increased temperatures, a different drive might be the total opposite. It's besides been proven past PC Perspective that the drive'south controller will throttle if it becomes too hot, so I would in theory want to cool down the controller, but heat the NAND chips to get optimal functioning.

Just like the SSD 840 EVO, there are OEM variants of the standard SSD 840. The Samsung SSD PM841 uses the aforementioned 21nm NAND, as does the SSD PM843. Samsung claims these drives do non have the speed degradation issue either, but the data higher up speaks for itself. We have tested a second SSD 840 in-business firm that shows the same degradation patterns forth with the numerous reports that can be found online.

Dell is not the only OEM using the drives either. The Razer Bract Pro laptop nosotros reviewed recently sports one of these Samsung OEM drives, and honestly "like new" operation is really good. Microsoft also uses them in the Surface Pro 3, and Samsung went as far equally releasing a firmware update to fix read deposition, but to pull it later.

Yet another potentially affected product is the Samsung SSD 845DC EVO**, an enterprise drive for server use, specifically suited for and I hope you enjoy irony equally much as I do: " suitable for read-intensive applications ". The 845DC ECO uses NAND with the exact same office number as the 840 EVO and PM851: K90KGY8S7M-CCK0. Samsung may take binned the all-time part of the TLC NAND for use in this bulldoze, so the trouble is likely to have a bit longer to prove upwards on it.

Quoting Samsung below, these were the kind of claims the company made when it promoted the apply of TLC NAND. In all fairness, nosotros've recommended the drives in our reviews, as have many others, and we happen to be using a handful of them in our systems too.

"To summit it all off, you tin residuum bodacious knowing that your SSD will continue to offer fantabulous performance throughout its useful lifespan. With its simple upgrade solution and sustained, industry- leading performance, the Samsung SSD 840 is the unmarried best upgrade you can make to your PC."
Source

"The 840 Series represents the first consumer SSD to implement 3-fleck/cell MLC (also called TLC) technology (...) This is goose egg a good firmware algorithm tin't handle, however. Samsung'southward iii-bit/prison cell MLC-based SSD 840 Series, equipped with mandatory OP, will all the same far outlast the useful life of the hardware it powers."
Source

Reliability, as in data loss, has not been put into question. So here's my open up request to Samsung: admit the problem exists in all the afflicted drives as evidenced in this article and in the countless reports institute in this lengthy thread on the Overclockers.net forums and elsewhere online.

As of writing, this single discussion has gathered over two,770 replies and 345,000 views. Thus far Samsung has decided to ignore the SSD 840 and all the aforementioned variants fifty-fifty though the drives carry 3-year warranties. Samsung, the ball is on your courtroom now...