Archive Home arrow Reviews: arrow Storage arrow OCZ RevoDrive PCI-Express SandForce SSD
OCZ RevoDrive PCI-Express SandForce SSD E-mail
Reviews - Featured Reviews: Storage
Written by Olin Coles   
Wednesday, 22 September 2010
Table of Contents: Page Index
OCZ RevoDrive PCI-Express SandForce SSD
Features and Specifications
Closer Look: OCZ RevoDrive PCI-E SSD
SandForce PCI-E SSD Anatomy
SSD Testing Methodology
AS-SSD Benchmark
ATTO Disk Benchmark
CrystalDiskMark 3.0 Tests
Iometer IOPS Performance
EVEREST Disk Benchmark
HDD vs Hybrid Drive vs SSD
OCZ RevoDrive Conclusion

Iometer IOPS Performance

Iometer is an I/O subsystem measurement and characterization tool for single and clustered systems. Iometer does for a computer's I/O subsystem what a dynamometer does for an engine: it measures performance under a controlled load. Iometer was originally developed by the Intel Corporation and formerly known as "Galileo". Intel has discontinued work on Iometer, and has gifted it to the Open Source Development Lab (OSDL).

Iometer is both a workload generator (that is, it performs I/O operations in order to stress the system) and a measurement tool (that is, it examines and records the performance of its I/O operations and their impact on the system). It can be configured to emulate the disk or network I/O load of any program or benchmark, or can be used to generate entirely synthetic I/O loads. It can generate and measure loads on single or multiple (networked) systems.

To measure random I/O response time as well as total I/O's per second, Iometer is set to use 4KB file size chunks over a 100% random sequential distribution at a queue depth of 32 outstanding I/O's per target. The tests are given a 50% read and 50% write distribution. While this pattern may not match traditional 'server' or 'workstation' profiles, it illustrates a single point of reference relative to our product field.

The chart below illustrates combined random read and write IOPS over a 120-second Iometer test phase, where highest I/O total is preferred:

Iometer_Random_4K-IOPS_30QD_Results.png

Since Iometer wasn't the first test we conducted on the OCZ RevoDrive (Everest was), it's surprising to see that 'dirty' NAND flash was still able to produce nearly 70,000 IOPS with 32 outstanding I/O's per target.

From the onset, SandForce SSDs clearly outperform the competition when tested which a larger queue depth. In our Iometer tests, which use 32 outstanding I/O's per target and a random 50/50 read/write distribution, only the 'unrestricted' SandForce SSDs approach 50,000 IOPS. OCZ's Agility 2 promises 10,000 IOPS each way, and yet it delivers 23,376 total - outperforming their specification once again . The SandForce-Driven SSDs demonstrate the highest IOPS performance we've ever seen on a consumer storage device, and the 'unlocked' firmware further extends the performance level to as high as 50,000 IOPS. Benchmark Reviews discusses this topic in more detail in our SandForce SF-1200 SSD Firmware Comparison article.

Drive Hardware

In our next section, we test linear read and write bandwidth performance and compare its speed against several other top storage products using EVEREST Disk Benchmark. Benchmark Reviews feels that linear tests are excellent for rating SSDs, however HDDs are put at a disadvantage with these tests whenever capacity is high.



 

Comments 

 
# RE: OCZ RevoDrive PCI-Express SandForce SSDfunkz 2010-09-22 01:52
Page 4 incorrectly states "Silicon Image defines their SiI3124 chip as a PCI-Express to SATA-3GB/s controller" when it's PCI-X as listed elsewhere in the review. The SiI3124 is ancient, strange that OCZ would choose to use it since the PCI-E has to be bridged to PCI-X first with the Pericom, when there are native PCI-E RAID controllers readily available, such as the SiI3132 for example. But why even bother using SATA on a card? Instead of this convoluted process of PCI-E to PCI-X to SATA RAID to NAND, how about a simple PCI-E NAND controller?
Report Comment
 
 
# Platform dependant?RealNeil 2010-09-22 02:45
I wonder if the drive would perform as well on a P55 mainboard?
They have certain bandwidth restrictions on the PCI-E bus with multiple PCI-E slots occupied concurrently. (X16 is divided into X8 times two)
I just wonder if this architecture would effect the output of the drive, and if the drive would effect the output of the Video Card?
Report Comment
 
 
# DESEMPENHO DOS DRIVES SSDSandro 2010-09-22 03:27
VEJA SO, SANDRO, QUE A INDUSTRIA TENTOU COLOCAR UM PREÇO ALTO NESTE TIPO DE "STORAGE DEVICE" MAS... NAO COLOU.
PARECE QUE VAO MUDAR A TATICA!
Report Comment
 
 
# RE: DESEMPENHO DOS DRIVES SSDRealNeil 2010-09-22 03:54
Did he say?: SO SEE, SANDRO, THAT TRIED TO PUT AN INDUSTRY'S HIGH PRE THIS KIND OF "STORAGE DEVICE" BUT ... Not getting through. IT SEEMS THAT WILL CHANGE THE TACTIC!
Report Comment
 
 
# TranslationGustavo 2010-09-22 22:50
What he said can roughly be translated as( not translated work by word, but a little better translation): "You see, Sandro,the industry tried to put a high price in this type of Storage device...but it didnt worked out. Now it look like they are going to change tatics."

Aside from the tranlation, can someone tell me if this type of storage( via Pci-e)is bootable??
Report Comment
 
 
# RE: TranslationRealNeil 2010-09-23 04:25
Ha-Ha! I used Google Translate on it and that's the best it would do for me. I don't understand Portuguese at all. You've made it much clearer to me what he meant. Thank You.
Report Comment
 
 
# RE: TranslationRealNeil 2010-09-23 04:30
As to the bootable, check OZC's site for more info on this device. It's bootable with some motherboards and is not with others.
Report Comment
 
 
# RE: TranslationAdam 2010-09-23 04:33
It is, although it's fairly dependant on the motherboard.

I'd guess quite a lot will need bios updates.
Report Comment
 
 
# RE: RE: TranslationOlin Coles 2010-09-23 06:55
Most current-generation motherboards will support Revo as a boot device, such as Intel's 5x series and AMD's 8xx series, but older motherboards may require a BIOS update.
Report Comment
 
 
# RE: OCZ RevoDrive PCI-Express SandForce SSDRobert17 2010-09-22 04:26
Nice review. Couple of questions: this drive is very reminiscent of the ANS-9010 product which I felt at the time was very forward-looking. Any chance that this could be THE direction SSD products will be headed by all manufacturers? (I'm guessing that you are provided a bit more insight on an average day)

I notice the idle power consumption on the data sheet provided by OCZ is 7W. Although still small, it seems to be much higher than other SSDs. Is this correct? If so, is it an artifact of having two controllers on the card or of being a PCI-E footprint? Any idea?
Report Comment
 
 
# AnswersOlin Coles 2010-09-22 07:50
@ funkz: Page 4 is actually correct, as I re-state Silicon Image's product description: siliconimage.com/products/product.aspx?pid=27
I agree with you about the controller; it works, but better options exist. Perhaps it was a cost benefit issue.

@ Neil: OCZ doesn't recommend motherboards with restrictive bandwidth on the PCI-Express bus, and the P55 motherboard would be a concern if SLI or CFX were used. If it's a single VGA with Revo on P55, I don't see a problem.

@ Robert: A standard SSD has a controller and NAND, while the Revo has twice the amount of NAND ICs, a RAID processor, and PCI-E bridge. This is where the extra power consumption comes from.
Report Comment
 
 
# Beware Crosshair IV ownersJethro 2010-09-22 08:31
Be careful about purchasing the Revodrive before checking compatibility on OCZ's site. The Crosshair IV and I think the EVGA X758 SLI board does not work with the Revodrive. ASUS has yet to fix their BIOS to enable the drive to work and I don't know about the EVGA.
Report Comment
 
 
# RE: Beware Crosshair IV ownersOlin Coles 2010-09-22 08:37
To be more specific, Jethro is referring to the OCZ RevoDrive as the computer's boot drive. It will still work as a secondary drive.
Report Comment
 
 
# liveonc.weebly.comJay 2010-09-23 08:52
Would have preferred if they offered a PCIe 2.0 x16 SSD with 4-8 SiI3124 in 16 RAID 0 or 8 RAID 1+0 even if it's "only" a 120GB SSD, to make most of the cheap PCI-X controller & most of the MLC-NAND memory using all 16 PCIe lanes.
Report Comment
 

Comments have been disabled by the administrator.

Search Benchmark Reviews
QNAP Network Storage Servers

Follow Benchmark Reviews on FacebookReceive Tweets from Benchmark Reviews on Twitter