Putting Theory to Practice: Understanding the SSD Performance Degradation Problem

Let’s look at the problem in the real world. You, me and our best friend have decided to start making SSDs. We buy up some NAND-flash and build a controller. The table below summarizes our drive’s characteristics:

  Our Hypothetical SSD
Page Size 4KB
Block Size 5 Pages (20KB)
Drive Size 1 Block (20KB
Read Speed 2 KB/s
Write Speed 1 KB/s

 

Through impressive marketing and your incredibly good looks we sell a drive. Our customer first goes to save a 4KB text file to his brand new SSD. The request comes down to our controller, which finds that all pages are empty, and allocates the first page to this text file.


Our SSD. The yellow boxes are empty pages

The user then goes and saves an 8KB JPEG. The request, once again, comes down to our controller, and fills the next two pages with the image.


The picture is 8KB and thus occupies two pages, which are thankfully empty

The OS reports that 60% of our drive is now full, which it is. Three of the five open pages are occupied with data and the remaining two pages are empty.

Now let’s say that the user goes back and deletes that original text file. This request doesn’t ever reach our controller, as far as our controller is concerned we’ve got three valid and two empty pages.

For our final write, the user wants to save a 12KB JPEG, that requires three 4KB pages to store. The OS knows that the first LBA, the one allocated to the 4KB text file, can be overwritten; so it tells our controller to overwrite that LBA as well as store the last 8KB of the image in our last available LBAs.

Now we have a problem once these requests get to our SSD controller. We’ve got three pages worth of write requests incoming, but only two pages free. Remember that the OS knows we have 12KB free, but on the drive only 8KB is actually free, 4KB is in use by an invalid page. We need to erase that page in order to complete the write request.


Uhoh, problem. We don't have enough empty pages.

Remember back to Flash 101, even though we have to erase just one page we can’t; you can’t erase pages, only blocks. We have to erase all of our data just to get rid of the invalid page, then write it all back again.

To do so we first read the entire block back into memory somewhere; if we’ve got a good controller we’ll just read it into an on-die cache (steps 1 and 2 below), if not hopefully there’s some off-die memory we can use as a scratch pad. With the block read, we can modify it, remove the invalid page and replace it with good data (steps 3 and 4). But we’ve only done that in memory somewhere, now we need to write it to flash. Since we’ve got all of our data in memory, we can erase the entire block in flash and write the new block (step 5).

Now let’s think about what’s just happened. As far as the OS is concerned we needed to write 12KB of data and it got written. Our SSD controller knows what really transpired however. In order to write that 12KB of data we had to first read 12KB then write an entire block, or 20KB.

Our SSD is quite slow, it can only write at 1KB/s and read at 2KB/s. Writing 12KB should have taken 12 seconds but since we had to read 12KB and then write 20KB the whole operation now took 26 seconds.

To the end user it would look like our write speed dropped from 1KB/s to 0.46KB/s, since it took us 26 seconds to write 12KB.

Are things starting to make sense now? This is why the Intel X25-M and other SSDs get slower the more you use them, and it’s also why the write speeds drop the most while the read speeds stay about the same. When writing to an empty page the SSD can write very quickly, but when writing to a page that already has data in it there’s additional overhead that must be dealt with thus reducing the write speeds.

The Blind SSD Free Space to the Rescue
POST A COMMENT

377 Comments

View All Comments

  • Franco1 - Wednesday, March 18, 2009 - link

    I've been waiting a long time for this review. It was certainly worth the wait! I would love to see some benchmarks with 2+ drives in RAID configurations via onboard and add-on controller cards. Maybe another follow up? Reply
  • Howard - Wednesday, March 18, 2009 - link

    Looks like the Vertex is the drive to get, especially once the user base expands a bit. Reply
  • MagicalMule - Wednesday, March 18, 2009 - link

    Thanks for the article. Everyone is critiquing grammar and all this nonsense it seems, but I really enjoyed the article.

    It was very thorough and very informative.

    Keep up the good work. =).
    Reply
  • futrtrubl - Wednesday, March 18, 2009 - link

    You missed out a VERY significant step that causes the greater part of the slowdown associated with your scenario. After the block is read out to cache the block has to be erased before it can be written to again and as you pointed out earlier an erase cycle, and thus the entire read/modify+erase/write cycle, takes a relatively LONG time, much longer than a simple read/modify/write.

    Edward
    Reply
  • DrKlahn - Wednesday, March 18, 2009 - link

    I've worked in IT for 15 years and have played with very fast arrays and know a fair amount about storage. 2 months ago I replaced my Raptor boot/gaming drive with a GSkill Titan. In day to day use I have no stuttering. The only stutter I have seen was while installing a large patch, surfing with multiple windows/tabs open and using Outlook. It wasn't even a second. I did align the partition, turned off drive indexing and defragmentation, and turned on caching. In day to day use it simply kills the Raptor. Games and applications load in a fraction of the time. Vista boot time has decreased dramatically.

    This isn't a case of purchase justification. If the drive was a dud I would have moved it to a secondary machine, reinstalled the Raptor, and chalked it up as a bad decision. I simply have not run into any scenario in daily use that it performs worse than the drive it replaced and I have not seen any real stuttering in daily use.
    Reply
  • Gary Key - Wednesday, March 18, 2009 - link

    I have a GSkill Titan drive also and really like it. However, my experiences while positive overall, do not compare with yours when it comes to stuttering (yes, all optimizations have been done to the drive and OS). I still have significant stuttering problems when using multiple IM programs and having multiple windows/tabs open at the same time. I literally have to wait a few seconds when texting colleagues if more than two conversations are occurring at the same time as the system pauses, hitches, and stutters in this scenario. It is especially aggravating when on Skype and trying to text, speak, and transfer files at the same time. This does not occur on the Intel drive in my testing. Apparently, it is no longer a problem on the OCZ Vertex or Summit drives. Except for my example above, I would certainly use the Titan drive over my Raptor any day of the week. Reply
  • druc0017 - Wednesday, March 18, 2009 - link

    great article, keep up the good work, cant wait to see more updates, thx Reply
  • mikeblas - Wednesday, March 18, 2009 - link

    Is the Velociraptor really "World's fastest hard drive", as this article states? Faster than the Hitachi SAS drives? Reply
  • Gary Key - Wednesday, March 18, 2009 - link

    We have changed those statements to "...fastest consumer desktop hard drive...", that was the original intent of the statement, just clarified now. :) Reply
  • 7Enigma - Wednesday, March 18, 2009 - link

    I think the majority of us understood that. People just like to nit-pick. Reply

Log in

Don't have an account? Sign up now