Hellhammer
AnandTech Emeritus
- Apr 25, 2011
- 701
- 4
- 81
The 2nd article linked seems to state that while it recovers, it only recovers to half of its' original performance in random writes (unlike OCZ which does eventually recover to 100% full speed).
Could be that Toshiba is just limiting the 4KB random write speed at high queue depths to avoid fragmenting the drive more. In other words, there's likely fairly aggressive GC going on if you're writing that kind of data to the drive, which slows down the host IO requests.
Just a theory that came to my mind. Shouldn't be a big deal because consumer workloads usually stay within QDs of 1 and 5.
By the way, have you (or are you) going to get a chance to review this series of drives in the future?
The Strontium Hawk I reviewed (linked above) is a rebranded Toshiba drive (even has the Toshiba sticker on the back), so I'm not sure there's a need to review the same drive once. If Toshiba releases new drives, then I'll definitely try to get my hands on one.