Assistance - Calling all Samsung EVO owners
- Thread starter Aurhinius
- Start date
Its possible there is a bug or some issue with accessing the filesystem more directly via a location rather than opening a handle to an individual file - as I see no slowdown in normal file operations as a program would work. If you write a file of random data to free space or open an existing file specifically and read it back to benchmark it works at expected speeds.
Last edited:
I don't deny that performance seems fine on the drive.
Mine boots fine, loads things fine and plays fine but I'm using rapid which would mask the issue anyway to some degree.
It's not just a benchmark issue though as I have said I see the performance issue when attempting an acronis backup of the drive.
So there is an issue there just not entirely sure what.
It may well be that the drive hand off between dram cache and drive is bad with small files.
It stands to reason that the first section of my drive will have the OS installed on it which will be packed full of tiny files.
Which may well be the issue that the drive read speed for lots of small files is really bad (by design or bug) and that larger files don't exhibit this issue.
Throwing idea's out there.
Please test your drive regardless of whether you think it is fine though. It's not a witch hunt it's trying to understand an issue we have identified and judging how wide an issue it is.
Takes 2 mins to download and run the HDTune benchmark and it's free.
Ultimately a drive's read performance should not be effected by age nor file location on the drive. The drive should be able to read a data block on any nand cell at the same speed regardless of age but as this proves it just isn't happening. i.e reading cell 200 on chip 1 should be equally as fast as cell 1 on chip 4.
Locating the nand cell and then reading the data out should be the same.
As john and I have show we've tested sequential read's here so we are asking it to read cell1,2,3,4 in order which should be faster than asking for random cells but again it's not. It may well be how the data is distributed over cell boundaries or chip boundaries I don't know.
So although the file is sequential on the drive virtually it's not physically sequential and data could be distributed across chips I believe this is how the internal firmware file map would work. You would expect it to write a file across multiple chips to improve read access time as the controller has multiple channels. So faster to read all channels in one cycle rather than having to cycle the same chip to read out of it multiple times.
That may well explain the issue it may well be that samsung's attempt to limit drive wear is forcing small amounts of data to be less well distributed across the chips.
The more I type the more I think this makes sense with my limited understanding of nand and SSD technology.
Mine boots fine, loads things fine and plays fine but I'm using rapid which would mask the issue anyway to some degree.
It's not just a benchmark issue though as I have said I see the performance issue when attempting an acronis backup of the drive.
So there is an issue there just not entirely sure what.
It may well be that the drive hand off between dram cache and drive is bad with small files.
It stands to reason that the first section of my drive will have the OS installed on it which will be packed full of tiny files.
Which may well be the issue that the drive read speed for lots of small files is really bad (by design or bug) and that larger files don't exhibit this issue.
Throwing idea's out there.
Please test your drive regardless of whether you think it is fine though. It's not a witch hunt it's trying to understand an issue we have identified and judging how wide an issue it is.
Takes 2 mins to download and run the HDTune benchmark and it's free.
Ultimately a drive's read performance should not be effected by age nor file location on the drive. The drive should be able to read a data block on any nand cell at the same speed regardless of age but as this proves it just isn't happening. i.e reading cell 200 on chip 1 should be equally as fast as cell 1 on chip 4.
Locating the nand cell and then reading the data out should be the same.
As john and I have show we've tested sequential read's here so we are asking it to read cell1,2,3,4 in order which should be faster than asking for random cells but again it's not. It may well be how the data is distributed over cell boundaries or chip boundaries I don't know.
So although the file is sequential on the drive virtually it's not physically sequential and data could be distributed across chips I believe this is how the internal firmware file map would work. You would expect it to write a file across multiple chips to improve read access time as the controller has multiple channels. So faster to read all channels in one cycle rather than having to cycle the same chip to read out of it multiple times.
That may well explain the issue it may well be that samsung's attempt to limit drive wear is forcing small amounts of data to be less well distributed across the chips.
The more I type the more I think this makes sense with my limited understanding of nand and SSD technology.
Last edited:
Here's some from my 250GB Evo:
I'm running a very old Q6600-based machine
so it maxes out at SATA2 speeds. My poor read performance is spread more 'widely' across the drive - not just the 'start' (if these terms mean anything on an SSD... 
)
I'm running a very old Q6600-based machine
so it maxes out at SATA2 speeds. My poor read performance is spread more 'widely' across the drive - not just the 'start' (if these terms mean anything on an SSD... )
Yeah Selekt0r you certainly are seeing the nasty sub 100MB for a good part of your drive.
The 2.7MB trough looks a bit dramatic as well.
Definitely a pattern here and I think it's perhaps a dirty reality of TLC nand but only shows up in certain benchmarks.
No word from Samsung so far.
The 2.7MB trough looks a bit dramatic as well.
Definitely a pattern here and I think it's perhaps a dirty reality of TLC nand but only shows up in certain benchmarks.
No word from Samsung so far.
Here's mine i cant use HD Tune, because its expired. But i had a feeling the trim or something wasnt working, because on a big database lookup i was doing, performance monitor in windows was showing reads of approx 25-28MB/sec
Checking with HD speed i see the same. With the drive starting off around 400MB/sec and after 1-2% it drops to around 25MB.
EVO 840 120GB
Checking with HD speed i see the same. With the drive starting off around 400MB/sec and after 1-2% it drops to around 25MB.
EVO 840 120GB
Mine was purchased in August 2013 and according to the SMART info has been powered on for approx. 7700 hours and approx. 8.5TB written. TRIM is enabled, and running the 'Optimise Performance' in Samsung Magician has made no difference.
Ordered mine on Thu 6th Mar 2014, It says TRIM is enabled and working, same for SATA3. Using the optimise performance does nothing of note.
Smart says 1.46TB Written and 3526 power on hours (it's a 24/7 media center).
I actually bought it to replace my garbage crucial SSD drive, which had to be disconnected from the motherboard once a month for 6 hours to get its performance back, otherwise it would stall windows. As it was the C (OS) drive for my media center, im afraid i couldnt be putting up with that crap.
So no more Crucial SSD's. No more Samsung SSD's. I think its back to Intel SSD's only in the future.
Last edited:
Toytown that is pretty dramatic and intersting given your detected issues with DB performance.
Once again showing this isn't a benchmark only issue.
Please keep posting folks we need more people with more EVO drives to run this very quick benchmark.
Once again showing this isn't a benchmark only issue.
Please keep posting folks we need more people with more EVO drives to run this very quick benchmark.
I have definitely noticed that my PC feels less 'snappy' recently in regular usage recently (I know this is subjective). If I happen to have some (even small) Windows system files being read at sub-10MB/s then this isn't very surprising...
Associate
- Joined
- 20 Apr 2014
- Posts
- 1,865
- Location
- Herts/uk
my evo
can you see the pic?
can you see the pic?
Last edited:
can you see the pic?
Have to click through to imageshack and download it, but yes. Your one is overall even worse than mine (although the minimum not quite so bad).
An urgent response is needed from Samsung on this!
Here is the performance whilst reading from my MySql (just doing a sequential scan on a 3GB DB), cpu during this time is approx 1-4%. The EVO is reading anywhere between 24-30MB/sec
The drive at this point doesn't lock up windows, however it does stall other apps. Every hour on the same PC, i have an small app i made, read a webpage and write a single field to a txt file. This normally takes between 1-2 seconds to complete, whilst the DB is being read, it can take anywhere between 15-35mins.
Can someone explain what's going on with this here? I'm no expert on these things but I assume this is bad with the inconsistency? My drive is ~2 weeks old, have just over 60GB of data used, and is the only drive I'm currently using.
A comparison of my 840 Evo 1TB and 830 256GB.
Both are OS drives.
The 840 is in my gaming PC so isn't really doing anything at idle, the 830 is in my on 24/7 PC so will be running other things while doing the benchmarks.
The first HD Tach benchmark is a quick bench and the second a long bench.
The 840 is rather jittery over the first 25% of the HD Tune and HD Tach benchmarks.
Both are OS drives.
The 840 is in my gaming PC so isn't really doing anything at idle, the 830 is in my on 24/7 PC so will be running other things while doing the benchmarks.
The first HD Tach benchmark is a quick bench and the second a long bench.
The 840 is rather jittery over the first 25% of the HD Tune and HD Tach benchmarks.
Last edited: