Crucial m4 And Intel SSD 320: The Other SSD Competitors

Cost Of More Space, m4's Over-Provisioning

It might surprise you, but Crucial's Marvell-based drives don't have any over-provisioning. The end result is more user accessible space, but it's possible that this may impact performance.


Crucial m4 256 GB
Intel SSD 510 256 GB
NAND Flash Components
25 nm MLC, ONFI 2.234 nm MLC, ONFI 2.1
Raw NAND
256 GB
256 GB
IDEMA Capacity
256 GB250 GB
Over-Provisioning
0%
3.4%
Windows Capacity
238.5 GiB232.8 GiB
Interface
SATA 6Gb/sSATA 6Gb/s


Over-Provisioning, Garbage Collection and Performance

Over-provisioning refers to space set aside so that a drive can perform bookkeeping functions. As you start using a SSD, collecting scattered blocks helps allocate space for future writes. This is what we know as "garbage collection."

Zero percent over-provisioning might cause the drive to perform slower as it fills up. In our case, we're just copying entire directories from our boot drive until the SSD is full. If there is a problem, performance will fall after we've filled the drive.

HD Tach RW
Clean Performance
Drive Full
After Idling 30 minutes
Intel SSD 510 (250 GB)Average Read: 370.8 MB/s
Average Write: 300.4 MB/s
Average Read: 371.1 MB/s
Average Write: 221.0 MB/s
Average Read: 339.4 MB/s
Average Write: 274.3 MB/s
Crucial m4 (256 GB)Average Read: 391.2 MB/s
Average Write: 233.8 MB/s
Average Read: 177.1 MB/s
Average Write: 253.6 MB/s
Average Read: 156.2 MB/s
Average Write: 253.9  MB/s


As expected, we see some performance degrade when we pack the drives full of data. After we wait 30 minutes to let idle garbage collection kick in, we see most of the performance recover, but Crucial's m4 still falls a bit short from what you would get in a out-of-box state. Specifically, look at how much read speeds suffer.

Intel SSD 510: Clean PerformanceIntel SSD 510: Clean PerformanceIntel SSD 510: Drive Full PerformanceIntel SSD 510: Drive Full PerformanceIntel SSD 510: After 30 minutesIntel SSD 510: After 30 minutes

Crucial m4: Clean PerformanceCrucial m4: Clean PerformanceCrucial m4: Drive Full PerformanceCrucial m4: Drive Full PerformanceCrucial m4: After 30 minutesCrucial m4: After 30 minutes

Intel's SSD 510 read speed only falls about 8.5% and we see write speed recover. In comparison, the m4's read speed falls about 40%. Write performance is better, but there are many peaks values. Clearly, garbage collection on the 510 provides higher and more consistent performance that we don't get with the m4. In this specific scenario, Crucial's latest drive seems to back itself into a corner.

Now it's also possible to test another aspect of garbage collection. Specifically, we can see how fast blocks become available. This involves looking at the performance of a drive after TRIM. Why? Remember the TRIM command doesn't erase all the data in the same way a secure erase does. It only tells the controller that the OS no longer occupies the storage space. This means you have to erase a block in order to make it available to any subsequent write event. Herein lies the contradiction. Individual blocks can't be erased, only groups of large segments can. So once a drive is completely filled up and TRIMed, the write speed is effectively limited by the speed at which a controller recycles.

We can do this by filling the drive again. Then, we empty the drive by using the Recycle Bin to trigger the TRIM command. If the sequential write performance after TRIM is less than the performance get get with a secure erase, it means that there is a bottleneck in the recycling process.

AS SSD Sequential Write Performance
Secure Erase Performance
After TRIM
Intel SSD 510 (250 GB)

315.75 MB/s

308.86 MB/s

Crucial m4 (256 GB)

283.12 MB/s

279.36 MB/s


There is some performance loss after issuing a TRIM command. Crucial's m4 writes sequential data at 283.12 MB/s in its clean state. After filling and emptying the m4, we find that sequential write performance is now 279.36 MB/s. With a 3% drop, we don't have to worry about the m4's performance being handicapped. When it comes to over-provisioning, we don't see a bottleneck in the recycling pathway.

This thread is closed for comments
33 comments
    Your comment
  • rainwilds
    Oooo, Crucial or Vertex? Decisions, decisions!
  • Could you expand on the Full Disk Encryption capabilities of the Intel 320?
    If you can actually use hardware FDE on that drive (rather than just secure erase), that's a winner for me.
  • bto
    Why does the Intel 510 250GB appear to have two scores in crystalmark? (469.4 and 259.7) on the top benchmark on page: "Benchmark Results: CrystalDiskMark Streaming Performance" the specs are identical for both.
  • poppasmurf
    Great lil tidbit, wonder what the difference will be between other SSD's with different interface connections other than physical appearance and the interface connection. More on the lines of pro's and con's between the SSD interface connections I'm referring to the OCZ PCI-e drives vs. SATA 6GB just a thought to stir up the hoop la of ssd's :P
  • JohnnyLucky
    I am beginning to wonder if we haven't reached the point where the human eye and brain are finding it harder to differentiate performance among ssd's. Some mainstream benchmarks seem to suggest that. Some of the benchmarks in this review seem to indicate the same. There are some very tight groupings.
  • henryvalz
    At the speed points that SSDs are functioning, I'm beginning to think that durability and reliability might be the best basis for decision. I would also really like to see some boot times from Windows 7, or loading time for games.
  • kev_stev
    Does anyone know when the vertex 3 and M4 are going to actually be available? I have heard rumors that the vertex 3 will be released "any day now" since mid march...
  • iamtheking123
    I'll go SSD in my next build, probably in a year and a half. Right now I'm satisfied with Raid 0-ed 1TB Caviar Blacks.
  • foscooter
    No mention of a release date. When will they be "in stores?" Q2 isn't exact enough.
  • zerapio
    Alert! Spelling police is coming and their PISSED

    (yes, it was intentional)
  • microking4u
    Why are the I/O's for this drive way off on your review compared to others such as Anand and PCPer?
  • groberts101
    Would have been interesting to see those Vantage marks on a Vertex 3 that hadn't already been hammered into a throttled state by all the previous tests. While it obviously shows the stamina and expected performance of the V3 after extremely heavy usage, the test doesn't take into consideration what many will see on newly installed drives that are used moderately. From that standpoint, the testing protocol is flawed, IMO.

    IOW, the testing protocol in reverse would have been more interesting to see typical Vantage scores from an unthrottled controller. I know for fact through personal beta-testing of the V3 that they would have been much higher.

    Or even better yet would be too take into account the special Durawrite throttling which the Sandforce drives STILL have built into the firmware(though not nearly as aggressive as the V2). Would surely give a nice little boost to SF through secure erase cleansing. If done at the 50% point it would show the potential in certain portions of the test suite that most WOULD see when not hitting thier drives with benchmark after benchmark in some sort of "hammer em' till the dust settles" protocol.

    Decent enough writeup though and all the review sites will eventually get it figured out, I guess.
  • PraxGTI
    Our SQL server has done more than 5*10^25 I/O Write Bytes in its 3 years lifespan. I really doubt the reliability of SSDs in a corporate server, although the IOs would be nice since our server can be crippled to 500% disk usage with disk queue sizes up to 8 at times.
  • praxgtiOur SQL server has done more than 5*10^25 I/O Write Bytes in its 3 years lifespan. I really doubt the reliability of SSDs in a corporate server, although the IOs would be nice since our server can be crippled to 500% disk usage with disk queue sizes up to 8 at times.


    How did you work that one out,

    10^24 bytes is a 1 yobibyte = 2^80 bytes = 1208925819614629174706176 bytes = 1,024 zebibytes

    1 zebibyte = 270 bytes = 1180591620717411303424 bytes = 1,024 exbibytes

    1 exbibyte = 260 bytes = 1152921504606846976 bytes = 1,024 pebibytes

    All of the data in the world on every hard drive is estimated at around 500 exbibytes.

    even in bits you are in order of several magnitude off
  • acku
    342145 said:
    Would have been interesting to see those Vantage marks on a Vertex 3 that hadn't already been hammered into a throttled state by all the previous tests. While it obviously shows the stamina and expected performance of the V3 after extremely heavy usage, the test doesn't take into consideration what many will see on newly installed drives that are used moderately. From that standpoint, the testing protocol is flawed, IMO. IOW, the testing protocol in reverse would have been more interesting to see typical Vantage scores from an unthrottled controller. I know for fact through personal beta-testing of the V3 that they would have been much higher. Or even better yet would be too take into account the special Durawrite throttling which the Sandforce drives STILL have built into the firmware(though not nearly as aggressive as the V2). Would surely give a nice little boost to SF through secure erase cleansing. If done at the 50% point it would show the potential in certain portions of the test suite that most WOULD see when not hitting thier drives with benchmark after benchmark in some sort of "hammer em' till the dust settles" protocol. Decent enough writeup though and all the review sites will eventually get it figured out, I guess.


    Hi groberts101,

    The test are actually run backwards. We just have help in a different order in the review. :)

    Cheers,
    Andrew
    TomsHardware
  • acku
    433483 said:
    Why does the Intel 510 250GB appear to have two scores in crystalmark? (469.4 and 259.7) on the top benchmark on page: "Benchmark Results: CrystalDiskMark Streaming Performance" the specs are identical for both.


    I think there is a legend in the lower right hand corner. One is using the 6Gb/s port and one is using the 3Gb/s port.
  • acku
    254934 said:
    Why are the I/O's for this drive way off on your review compared to others such as Anand and PCPer?

    Which ones are you referencing?
  • ww2003
    From what i have been hearing the new vortec 3 is going to be the best SSD on the market with faster speeds the any other one has right now.
  • zodiacfml
    I like the part in the conclusion that one not need the fastest SSDs out there especially for desktop uses.
    In my opinion, Intel has a point with their new products and pricing, enable bigger capacities at lower capacities.
  • zodiacfml
    I meant prices. :p

    zodiacfmlI like the part in the conclusion that one not need the fastest SSDs out there especially for desktop uses. In my opinion, Intel has a point with their new products and pricing, enable bigger capacities at lower capacities.
  • vvhocare5
    A couple of observations:
    1. I was showing these charts to a coworker and pointed out that the worst SSD had 10x better performance than a good current rotating hard drive - and the good ones were 50x or better than that. We can split hairs over which SSD but the question I ask is why are you still on a hard drive?

    2. The stupid SQL server calc continues to show the ignorance people have about SSD's. 99% of that SQL server's use is READS. SSD's can read all day long - its the write cycles that count against them. The early Intel drives guaranteed 100G of writes per day for 5 years - Im not sure if the other mfg list these specs or if they are unchanged.

    In business use the question you have to ask yourself is does the cost of an SSD drive either defer an upgrade or somehow return more money than you spent on it. If no, then dont buy SSD's. For personal use we have no return - its pure fun and enjoyment to see how fast these things are....
  • You show that the read performance of the m4 degrades significantly over time with usage. How did this degradation affect the benchmark performance of the drive? I'm mostly interested in the PCMark Vantage benchmarks where the m4 was pretty consistently the leader -- was that a fresh drive or a used drive?
  • "5*10^25 I/O Write Bytes in its 3 years lifespan" seems to be a bit far fetched...

    5*10^25 =
    50,000,000,000,000,000,000,000,000 Bytes

    Number of seconds in 3 years: 3*365*24*60*60 = 94,608,000 sec

    Divide bytes by seconds to get continuous average transfer rate:
    528,496,533,062,743,108 Bytes/sec

    Roughly 7 Billion people in the world (assuming everyone has internet which they dont, and everyone is always downloading, which they aren't)

    Transfer per person:
    75,499,504 Bytes/sec/person

    Which is 72MB per person per second. Not possible!

    This would mean 7 Billion people in the world are consuming data constantly with their 578 Mbs (mega-bit per second) Internet connection!!

    I WANT A 578 Mbs internet connection!!!
  • acku
    Quote:
    You show that the read performance of the m4 degrades significantly over time with usage. How did this degradation affect the benchmark performance of the drive? I'm mostly interested in the PCMark Vantage benchmarks where the m4 was pretty consistently the leader -- was that a fresh drive or a used drive?

    Fresh out of the box. You'll see the m4 hit 65k in the PCMark HDD suite. I'll have to double check the numbers, but off the top of my head I think it was like ~54k in a steady state. Don't quote me on that though.

    Remember this is for the 256 GB drive. The m4 128 GB model is suppose to hit around 55k right out of the box.