Random Read Performance
The random read test requests 4kB blocks and tests queue depths ranging from 1 to 32. The queue depth is doubled every three minutes, for a total test duration of 18 minutes. The test spans the entire drive, which is filled before the test starts. The primary score we report is an average of performances at queue depths 1, 2 and 4, as client usage typically consists mostly of low queue depth operations.
The random read performance of the CS1311 is surprising. The 120GB outperforms most other planar TLC drives, and the larger capacities clearly outperform the OCZ Trion 150. The performance of the CS2211 is mediocre by MLC standards but still significantly better than the planar TLC drives.
The PNY drives show a clear pattern to the power consumption: reading from more flash chips takes more power, and since read operations aren't much more power-hungry on TLC than MLC, the higher performance of MLC leads to higher power consumption.
 | |||||||||
The 120GB CS1311 scales better in both power and performance than its larger counterparts, while the relation between the 240GB and 480GB is more typical. The CS2211 demonstrates significantly higher power consumption for the 480GB across the board for little performance benefit.
Random Write Performance
The random write test writes 4kB blocks and tests queue depths ranging from 1 to 32. The queue depth is doubled every three minutes, for a total test duration of 18 minutes. The test is limited to a 16GB portion of the drive, and the drive is empty save for the 16GB test file. The primary score we report is an average of performances at queue depths 1, 2 and 4, as client usage typically consists mostly of low queue depth operations.
On the random write test, the 120GB CS1311 again shows higher performance than its higher capacity counterparts, but by a very slim margin. The CS2211s are quite close in performance to each other and to other mid-range MLC drives. PNY's MLC drive sustains more than twice the throughput of their TLC drive.
Power consumption during random writes is very similar across all the PNY drives, showing that the higher per-operation power cost on TLC is roughly matched by the higher throughput of MLC.
 | |||||||||
The CS1311 shows little to no performance scaling with higher queue depths, although power consumption does increase slightly. The CS2211 plateaus once the queue depth reaches eight.
ncG1vNJzZmivp6x7orrAp5utnZOde6S7zGiqoaenZH5xfpJtZq2glWK9r8WMnKpqa2FmeqK6w2aarGpiZn5uv9KdZKudpp6yuHnMpZpmrqNiwa2vjJqrZmllo7pwgg%3D%3D