Jump to content


Photo

Does Kingston's reasoning behind my slow write speed performance m


  • You cannot start a new topic
  • Please log in to reply
6 replies to this topic

#1 Matrix Leader

Matrix Leader

    Member

  • Member
  • 2 posts

Posted 26 April 2013 - 06:54 AM

I bought a Kingston HyperX 3K SSD 480 GB

I feel my OS is snappy just as it should on an SSD but in my Benchmarks on CrystalDisk Mark and AS SSD Benchmark, I got 500 MB/S read which is good and 170 MB/S write only

Now the advertised speed of the SSD on the box says 540 MB/S Read and 450 MB/S write

So as we can see here, my write speeds are in no way near what is advertised.

I then upgraded the firmware from the original firmware 501 to 503 and that made my write speeds double from 170 to 316 MB/S which is better, but still no where close to the advertised speed.

Upon contacting them, they advised me to run the ATTO Disk Benchmark and then I got 530 MB/S read and 450 MB/S write as advertised; reason being, is that the ATTO Disk Benchmark they say uses compression in the benchmarks which is where this drive excels and the other benchmarks such as CrystalDisk Mark and AS SSD Benchmark don't use compression.

That is all well and good but my question is, in my real world usage, does Windows or any other program for that matter use compression? I mean, how does their 450 MB/S write speed in compression mode work in real life? or is it just a benchmark gimmick because when copying files from the SSD onto itself, example:

I try to copy a 16 GB movie from (which the 2nd partition in this SSD) onto another location in

The write speed I see in Windows goes from 420 MB/S at first, drops to 370 MB/S within 2 seconds, then keeps dropping gradually all the way until it reaches a stable 135 MB/S

This sound very crappy to me and I think that compression in ATTO Disk Benchmarks is just a gimmick because who uses compression??

Hope someone can shed some light on the above

Thanks

#2 rugger

rugger

    Member

  • Member
  • 724 posts

Posted 28 April 2013 - 02:00 AM

Your performance is quite normal in my experience with the 480gig Sandforce drives. The fastest sandforce drives are definitely the 240gig drives.

Movies are already highly compressed, and will always show the worst case write performance for the sandforce drives.

Your programs and settings will usually be in a more compressible format, allowing for faster performance. Not as fast as the ATTO results, but not as slow as copying movie files. The compression that sandforce drives use is typically better for enterprise workloads like databases though, where the compressible nature of many of the writes greatly increases performance and lifetime of the drives.

To answer your other thread, over-provisioning won't generally help you. Just keep the drive less then 80% full and you won't see the any performance problems.

Edited by rugger, 28 April 2013 - 02:04 AM.

#3 [ETA]MrSpadge

[ETA]MrSpadge

    Member

  • Member
  • 764 posts

Posted 29 April 2013 - 04:10 AM

Rugger is right, it'd just like to add: it's not about "who uses compression" but rather "which data is compressible", as the Sandforce controller compresses everything in hardware. With regular data (OS, programs, games etc.) this works well, but not with already highly compressed stuff like movies. Well, with mpgs there might still be some gain :D

MrS

#4 Beenthere

Beenthere

    Member

  • Member
  • 178 posts

Posted 29 April 2013 - 11:29 AM

Matrix-

If it's any consolation you could take the fastest SSD and the slowest SSD and hardly see any performance difference between them in daily use for a typical PC.

The advertised benchmark performances are used to sell the drives. If the SSD maker uses proper disclosure they will say that the results are on a new drive with no other data or programs installed and that the results are the best possible and not necessarily the typical results that consumers are likely to see. In normal use the benches would be lower but overall the speed improvement is significant over a HDD.

#5 [ETA]MrSpadge

[ETA]MrSpadge

    Member

  • Member
  • 764 posts

Posted 29 April 2013 - 04:19 PM

Beenthere.. that's a correct but very general statement. I his case the (main part of the) discrepancy is clearly caused by the special Sandforce real-time de/compression, which makes the drives performance rely on the type of data transmitted.

MrS

#6 MRFS

MRFS

    Member

  • Member
  • 190 posts

Posted 29 April 2013 - 06:12 PM

Ditto the excellent replies you've received already.

If you want to demonstrate for yourself the differences
between compressible and UNcompressible measurements
with a SandForce controller, download CrystalDiskMark.

Under the File menu, note that "Test Data" has
3 options:

Default(Random)
zero fill
one fill

So, run it once with "Default":
that option will use the LEAST
compressible data.

Then, run it 2 more times,
once with zero fill and
once with one fill.

A continuous stream of ALL ZEROES
or ALL ONES is the MOST compressible.

You should notice a big difference
between Default and either of the
other options, because all zeroes
and all ones are the MOST compressible,
and Default (Random) is the LEAST compressible.


With other, more modern SSD controllers,
like those in the Samsung 840 Pro,
you should NOT see such large
differences between compressible
and UNcompressible data.


Hope this helps.


MRFS

#7 MRFS

MRFS

    Member

  • Member
  • 190 posts

Posted 29 April 2013 - 06:23 PM

> compression in ATTO Disk Benchmarks is just a gimmick because who uses compression??

It's not a "gimmick" so much as a deliberate feature
of ATTO that helps SandForce controllers LOOK GOOD.

Another feature of ATTO you should try is "Direct I/O":
this feature bypasses the associative buffers
in the NTFS file system, in order to obtain
a controlled measurement of the raw hardware --
by eliminating the effects of OS file system buffering.

It's also very educational to UNCHECK "Direct I/O"
because there are probably very few application programs
that will bypass the NFTS buffers during normal production
usage e.g. Internet browsing, gaming etc.

For that reason, one should run ATTO twice:
once with "Direct I/O" enabled, and
once with "Direct I/O" disabled.

The latter option is actually the more "realistic"
measure of the maximum performance you can expect
from a SandForce controller, i.e. when the raw
data are compressible.

Even when the raw data are UNcompressible,
the performance effects of NTFS buffers
are substantial, particularly when a fast CPU
and fast DRAM are installed in your system.

This effect is very similar to the average performance improvement
one experiences by upgrading a rotating HDD with a 2MB cache
to one with a 64MB cache.


MRFS

Edited by MRFS, 29 April 2013 - 06:24 PM.




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users