Beyond the average data rate reported on the previous page, there's even more information we can collect from Tom's Hardware's Storage Bench. For instance, mean (average) service times show what responsiveness is like on an average I/O during the trace.
It would be difficult to graph the 10+ million I/Os that make up our test, so looking at the average time to service an I/O makes more sense. For a more nuanced idea of what's transpiring during the trace, we plot mean service times for reads against writes. That way, drives with better latency show up closer to the origin; lower numbers are better.
Write latency is simply the total time it takes an input or output operation to be issued by the host operating system, travel to the storage subsystem, commit to the storage device, and have the drive acknowledge the operation. Read latency is similar. The operating system asks the storage device for data stored in a certain location, the SSD reads that information, and then it's sent to the host. Modern computers are fast and SSDs are zippy, but there's still a significant amount of latency involved in a storage transaction.
We get some good news for Plextor's M6 family. Collectively, both of the 256 GB samples we're testing stand up to the older (and well-respected) M5 Pro. Considering the M6es are currently more mainstream-oriented, that's positive indeed.
Although the M5 Pro does fare better in reads, our write measurements are largely comparable. That's no small feat given the dynamics of 64 and 128 Gb dies at the 256 GB capacity point.
Mean Read Service Time
It could be the new flash, Marvell's controller, differences in firmware, or a combination of the three, but both M6-series SSDs finish just behind the M5 Pro. That's more good news, I'd say.
Mean Write Service Time
As I alluded to above, you don't get as much interleaving from 128 Gb dies at the 256 GB capacity point. Despite that reduction in parallelism, the M6es and M5 Pro achieve nearly the same mean write service times. That's fairly remarkable, since the 840 EVO, SP920, and M500 leverage the same number of dies and falls significantly behind Plextor's latest.
It's possible that Marvell's new controller gets some of the credit, though Plextor might also be more cleverly utilizing its NAND.
- Meet The M6S, M6M, And Another Marvell Controller
- How We're Testing Plextor's M6S And M6M
- Results: Random And Sequential Performance
- Results: Tom's Hardware Storage Bench v1.0
- Results: Tom's Hardware Storage Bench v1.0, Continued
- PCMark 8's Storage Consistency Test: New For Tom's Hardware
- Storage Consistency: The Adobe Photoshop (Heavy) Trace
- Results: Power Consumption
- Plextor Iterates Its Line-Up With New Components