OCZ Vector (256GB) Review
by Anand Lal Shimpi on November 27, 2012 9:10 PM ESTTimes are changing at OCZ. There's a new CEO at the helm, and the company is now focused on releasing fewer products but that have gone through more validation and testing than in years past. The hallmark aggressive nature that gave OCZ tremendous marketshare in the channel overstayed its welcome. The new OCZ is supposed to sincerely prioritize compatibility, reliability and general validation testing. Only time will tell if things have changed, but right off the bat there's a different aura surrounding my first encounter with OCZ's Vector SSD.
Gone are the handwritten notes that accompanied OCZ SSD samples in years past, replaced by a much more official looking letter:
The drive itself sees a brand new 7mm chassis. The aluminum colored enclosure features a new label. Only the bottom of the SSD looks familiar as the name, part number and other details are laid out in traditional OCZ fashion.
Under the hood the drive is all new. Vector uses the first home-grown SSD controller by OCZ. Although the Octane and Vertex 4 SSDs both used OCZ Indilinx branded silicon, they were both based on Marvell IP - the controller architecture was licensed, not designed in house. Vector on the other hand uses OCZ's brand new Barefoot 3 controller, designed entirely in-house.
Barefoot 3 is the result of three different teams all working together. OCZ's UK design team, staffed with engineers from the PLX acquisition, the Korea design team inherited after the Indilinx acquisition, and folks at OCZ proper in California all came together to bring Barefoot 3 and Vector to life.
The Barefoot 3 controller integrates an unnamed ARM Cortex core as well as an OCZ Aragon co-processor. OCZ isn't going into a lot of detail as to how these two cores interact or what they handle, but multi-core SoCs aren't anything new in the SSD space. A branded co-processor is a bit unusual, and I suspect that whatever is responsible for Vector's distinct performance has to do with this part of the SoC.
Architecturally, Barefoot 3 can talk to NAND across 8 parallel channels. The controller is paired with two DDR3L-1600 DRAMs, although there's a pad for a third DRAM for use in the case where parity is needed for ECC.
Hardware encryption is not presently supported, although OCZ tells us Barefoot 3 is more than fast enough to handle it should a customer demand the feature. Hardware encryption remains mostly unused and poorly executed on client drives, so its absence isn't too big of a deal in my opinion.
OCZ does its own NAND packaging, and as a result Vector is home to a sea of OCZ branded NAND devices. In reality you're looking at 25nm IMFT synchronous 2-bit-per-cell MLC NAND, just with an OCZ silkscreen on it. There's no NAND redundancy built in to the drive as OCZ is fairly comfortable with the error and failure rates at 25nm. The only spare area set aside is the same 6.8% we see on most client drives (e.g. a 256GB Vector offers 238GB usable space in Windows).
OCZ Vector | |||||
128GB | 256GB | 512GB | |||
Sequential Read | 550 MB/s | 550 MB/s | 550 MB/s | ||
Sequential Write | 400 MB/s | 530 MB/s | 530 MB/s | ||
Random Read | 90K IOPS | 100K IOPS | 100K IOPS | ||
Random Write | 95K IOPS | 95K IOPS | 95K IOPS | ||
Active Power Use | 2.25W | 2.25W | 2.25W | ||
Idle Power Use | 0.9W | 0.9W | 0.9W |
Regardless of capacity, OCZ is guaranteeing the Vector for up to 20GB of host writes per day for 5 years. The warranty on the Vector expires after 5 years or 36.5TB of writes, whichever comes first. As with most similar claims, the 20GB value is pretty conservative and based on a 4KB random write workload. With more realistic client workloads you can expect even more life out of the NAND.
Despite being built on a brand new SoC, there's a lot of firmware carryover from Vertex 4. Indeed if you look at the behavior of Vector, it is a lot like a much faster Vertex 4. OCZ does continue to use its performance mode that enables faster performance if less than 50% of the drive's capacity is used, however in practice OCZ seems to rely on it less than in the Vertex 4.
The design cycle for Vector is the longest OCZ has ever endured. It took OCZ 18 months to bring the Vector SSD to market, compared to less than 12 months for previous designs. The additional time was used not only to coordinate teams across the globe, but also to put Vector through more testing and validation than any previous OCZ SSD. It's impossible to guarantee a flawless drive, but doing considerably more testing can't hurt.
The Vector is available starting today in 128GB, 256GB and 512GB capacities. Pricing is directly comparable to Samsung's 840 Pro:
OCZ Vector Pricing (MSRP) | ||||||
64GB | 128GB | 256GB | 512GB | |||
OCZ Vector | - | $149.99 | $269.99 | $559.99 | ||
Samsung SSD 840 Pro | $99.99 | $149.99 | $269.99 | $599.99 |
OCZ is a bit more aggressive on its 512GB MSRP, otherwise it's very clear what OCZ views as Vector's immediate competition.
151 Comments
View All Comments
jeffrey - Tuesday, November 27, 2012 - link
Anand,I would love to have seen results using the 1.5 firmware for the 256GB Vertex 4. Going from 1.4 to 1.5 is non destructive. The inconsistency of graphs in other SSD reviews that included the 512GB Vertex 4 drive with 1.5 firmware and the 256GB Vertex 4 drive with 1.4 firmware drove me nuts.
When I saw the Barefoot 3 press release on Yahoo Finance, I immediately went to your site hoping to see the review. I was happy to see the article up, but when I saw your review sample was 256GB I feared you would not have updated the firmware on the Vertex 4 yet. Unfortunately, my fears were confirmed. I love your site, that's why I'm sharing my $.02 as a loyal reader.
Take care,
Jeffrey
Anand Lal Shimpi - Wednesday, November 28, 2012 - link
Some of the results are actually using the 1.5 firmware (IO consistency, steady state 4KB random write performance). We didn't notice a big performance difference between 1.4 and 1.5 which is why I didn't rerun on 1.5 for everything.Take care,
Anand
iwod - Tuesday, November 27, 2012 - link
Isn't this similar? Sandforce comes in, reached top speed in SATA 6Gbps, then other controller, Marvell, Barefoot managed to catch up. That is exactly what happen before with SATA 3Gbps Port. So in 2013 we would have controller and SSD all offering similar performance bottlenecked by its Port Speed.When are we going to see SATA Express that give us 20Gbps? We need those ASAP.
A5 - Wednesday, November 28, 2012 - link
SATA Express (on PCIe 3.0) will top out at 16 Gbps until PCIe 4.0 is out. This is the same bandwidth as single-channel DDR3-2133, by the way, so 16 Gbps should be plenty of performance for the next several years.extide - Wednesday, November 28, 2012 - link
Actually I believe Single Channel DDR-2133 is 16GiB a sec, not Gb a sec, so Sata Express is only ~1/8th the speed of single channel DDR3-2133jwilliams4200 - Wednesday, November 28, 2012 - link
It is good to see anandtech including results of performance consistency tests under a heavy write workload. However, there is a small or addition you should make for these results to be much more useful.You fill the SSDs up to 100% with sequential writes and I assume (I did not see a specification in your article) do 100% full-span 4KQD32 random writes. I agree that will give a good idea of worst-case performance, but unfortunately it does not give a good idea of how someone with that heavy a writeload would use these consumer SSDs.
Note that the consumer SSDs only have about 7% spare area reserved. However, if you overprovision them, some (all?) of them may make good use of the extra reserved space. The Intel S3700 only makes available 200GB / 264GiB of flash, which comes to 70.6% available, or 29.4% of the on-board flash is reserved as spare area.
What happens if you overprovision the Vector a similar amount? Or to take a round number, only use 80% of the available capacity of 256GB, which comes to just under 205GB.
I don't know how well the Vector uses the extra reserved space, but I do know that it makes a HUGE improvement on the 256GB Samsung 840 Pro. Below are some graphs of my own tests on the 840 Pro. I included graphs of Throughput vs. GB written, as well as latency vs. time. One the 80% graphs, I first wrote to all the sectors up to the 80% mark, then I did a 80% span 4KQD32 random write. On the 100% graphs, I did basically the same as anandtech did, filling up 100% of the LBAs then doing a 100% full-span 4KQD32 random write. Note that when the 840 Pro is only used up to 80%, it improves by a factor of about 4 in throughput, and about 15 in average latency (more than a 100 times improvement in max latency). It is approaching the performance of the Intel S3700. If I used 70% instead of 80% (to match the S3700), perhaps it would be even better.
Here are some links to my test data graphs:
http://i.imgur.com/MRZAM.png
http://i.imgur.com/Vvo1H.png
http://i.imgur.com/eYj7w.png
http://i.imgur.com/AMYoe.png
Ictus - Wednesday, November 28, 2012 - link
Just so I am clear, did you actually re create the partition utilitizing 80% of the space, or just keep the used space at 80% ?jwilliams4200 - Wednesday, November 28, 2012 - link
No partitions, no filesystems. I'm just writing to the raw device. In the 80% case, I just avoided writing to any LBAs higher than 80%.JellyRoll - Wednesday, November 28, 2012 - link
Excellent testing, very relevant, and thanks for sharing. How do you feel that the lack of TRIM in this type of testing affects the results? Do you feel that testing without a partition and TRIM would not provide an accurate depiction of real world performance?jwilliams4200 - Wednesday, November 28, 2012 - link
I just re-read your comment, and I thought perhaps you were asking about sequence of events instead of what I just answered you. The sequence is pretty much irrelevant since I did a secure erase before starting to write to the SSD.1) Secure erase SSD
2) Write to all LBAs up to 80%
3) 80% span 4KQD32 random write