Why you can trust Tom's Hardware Our expert reviewers spend hours testing and comparing products and services so you can choose the best for you. Find out more about how we test.
Mixed Workload And Steady State Testing
Our mixed workload testing is described in detail here, and our steady state tests are described here.
With minimal conditioning, the Vector 180 960GB falls to the middle of the performance chart when mixing sequential reads and writes. At higher queue depths, it recovers and performs much better compared to other drives.
The 4KB mixed workload test puts the Vector 180 960GB near the bottom of the chart. We have an explanation for this, though.
Sequential Steady State Performance
This chart illustrates steady state sequential performance in 10% increments, with 100% read on the left and 100% write on the right. The Vector 180 and SanDisk Extreme Pro 960GB stand out from a number of exceptional consumer SSDs.
I want to single out the 80% read mix and plot that data. Intel states this is representative of consumer workloads, and we agree with the company's analysis. Again, the Vector 180 and Extreme Pro enjoy large leads over the competition.
Random Write Steady State Performance
This is where things start getting dicey for the Vector 180. In short, the Vector 180's buffer flush causes some issues. On the 960GB model with 1GB of DRAM, the process takes roughly two seconds if the flash writes at around 500 MB/s. When a flush occurs, it limits available bandwidth for other operations. We observed the two-second flush roughly five times every 100 seconds as we abused the drive with 4KB random writes.
Granted, you won't be peppering the drive with 4KB writes under normal conditions. When we asked about our findings, OCZ's CTO, Daryl Lang, had this to say:
“Thank you for your inquiry in regards to the I/O behavior on the Vector 180 Series. What is being observed is a characteristic of the design of the drive itself and is a result of the firmware performing updates to its metadata mapping table and flushing the entire table out of DRAM and onto the NAND flash, during which I/O throughput is impacted for very brief periods. Our metadata management is done on a frequent basis to prevent failure modes related to bricked drives as a result of metadata corruption, which can potentially happen on other non PFM+ enabled SSDs as a result of unexpected power loss. This is observed to a greater extent on the larger drives (960GB) where there is more metadata to manage. While this phenomenon is observable in synthetic benchmarks, there is virtually no impact to typical client grade end-user applications and during real world use. With the Barefoot 3 based Vector 180 Series design, we strove to deliver the optimal balance of performance and reliability for our valued enthusiast and workstation customers.”
As a follow-up, we wanted to see if the table flush was based on workload or time. So, we doubled the block size from 4KB to 8KB in an effort to lighten the table's updates. In the end, we found the system is based on the passing of time with a rate of one update every 20 seconds.
