The SSD Review The Worlds Dedicated SSD Education and Review Resource | 
TEST BENCH AND PROTOCOL
Testing devices that push out 2.5GB and over 400,000 IOPS and reaching their highest performance is no easy task. Although we utilized the Test Bench shown here, we mixed and matched several configurations of motherboards over the course of a week to determine the best results.
In the end, the results displayed are the highest achieved through testing on any of three motherboards, the Gigabyte Z68x-UD7-B3, ASRock Z68 Extreme 4 and the Z68 Extreme 7 which arrived less than twelve hours to publishing.
In testing, our main objective is to obtain results as pure and as accurate as possible and we want to ensure that no anomalies slip through. Simply put, we want to provide you with the absolute best results the tested hardware can provide.
Repetition in testing is standard and, if necessary, we may conduct specific tests in Windows 7 safe mode to ensure the OS has little to no influence on the end result.
In order to validate and confirm our findings, testing is supported by industry accepted benchmark programs. All results are displayed through capture of the actual benchmark for better understanding of the testing process by the reader.
We would like to thank ASRock, Gigabyte, Corsair, MSI, OCZ, Fractal-Design, and Icy Dock for sponsoring components of our Test Bench.
SSD COMPRESSION AND TESTING FLUCTUATIONS
All SSDs are not created equal and this is as important a realization in the testing of enterprice PCI Express SSDs as any other. The OCZ Z-Drive R4 SSD relies on SandForce SF-2281 processors for its speed, the performance being part and parcel to SandForces use of compression techniques in storage whereas this is not seen in FUSION-IO products. This creates a bit of confusion when enthusiasts, and even representatives of large companies, test the drive with random data through benchmarking programs such as AS SSD and Crystal Diskmark (random data sample). The results seem to be lower than the listed specifications.
The results that we will see in the Z-R4, in testing incompressible data, actually present a false portrayal of the drives ability when compared to the io-D. This is why we are going to provide an extensive sample of benchmarks which will afford a clear indication of the strengths and weaknesses in both drives.
BENCHMARK SOFTWARE
Software used for testing today will include ATTO Disk Benchmark, Crystal DiskMark, AS SSD, Anvil Storage Utilities, IOMeter 2010, HDTune Pro along with FutureMark PCMark Vantage.
All do a great job of showing us the numbers that we want to see, or dont want to see in some cases, while PCMark Vantage x64 is an excellent program which recreates tests that mimic the average users activity, all the while providing a medium to measure each.
LET THE STORAGE WARS BEGIN!!
OCZ Z-DRIVE R4 C SERIES PCI EXPRESS 1.6TB SSD
FUSION-IO IODRIVE DUO PCI-EXPRESS 640GB SSD
Would be more fair if you compared it to the 1.28 TiB variant of the Duo.
Brilliant article none the less…
I agree totally but we have to work with the hardware that we have access to. As for the compliment of it being a brilliant article, totally unexpected and thank you very much.
“we have to work with the hardware that we have access to”
OK, so you have access to the fastest OCZ hardware but not the fastest FIO hardware…and this makes for a subjective test? Seriously? When do you expect to have current hardware for live testing? Any chance we might get actual server-side testing to show off the true strength of FIO hardware and, more importantly, software? Oh wait, then we couldn’t have a test could we…you’d need a competitor other than OCZ….
WE would be pleased to hear suggestions as to other sources of competition for the card. Thanks ahead.
not sure why the harsh response or love affair with Fusion. Why do you care which company? I say Fusion should put up the hardware like OCZ and do real life tests within applications. Both companies can put there money where their mouth is.
Harsh response? Love affair? I guess I am a bit confused because, well, myself, when I read an article I want to know the reviewers opinion and feelings just as much as I do facts alone. This is what keeps my interest.
It is also an important factor to help the reader. Take for instance the stuttering 602 controller a few years back. Can you imagine a reviewer saying, ” The drive stalls and stutters a bit but is still a large step up from a hard drive” Say it like it is I believe and hope our fellow members feel the same.
Thank you for taking the time to respond and expressing your feelings.
This is subjective because the z-drive r4 is cheaper than the FIO drive.
You find it subjective because one performs alot better yet is alot cheaper? How so?
Please update using the ioDrive 2 they just released, also
Finally SSDs start approaching the RAM speeds. Not very useful yet for the regular users without deep pockets. So meantime we still get 2-5 times of these speeds with software based QSoft RAMDrive and, thanks to DRAM price drop, at the $5/GB. While many even never heard of that but we are enjoying such crazy RAMDrives’ speeds for our apps for almost a quarter of century since the DOS times.
Suggestion to authors – test it just for fun. The latest QSoft Enterprise edition or one of Romex Software have a lot of nice functions allowing to auto backup the RAMdrive and load it back when you switch on and off your computer. Yes it might be lost if PC crash if you do not autobackup things, but this typically never happen, It is not for servers or critically important files but in its turn it is soooooo much faster and is eternal (well, lifetime guarantee) and no wearing off, TRIM, problems with compressed or 4K files.
I do not know while similar board designs of RAMDrives as these two SSDs are not making its road to the world (with just the battery backups) – they are so much faster t yet
nice review. I like both of these companies, but I think you need to compare the iodrive II DUO if you really want to compare head to head what each manufacture would put against another. Another addition that would be nice is to see performance of actual work like indexed data using lucene. What is the performance gain on these working enviornments.
We have been in contact with FusionIO during CES and just may be able to appease your request soon enough.
OCZ Z-R5 is also coming
Sorry, but thinking this through HP should donate some equipment or Dell as well as updating the cards and also add tests of real applications to see performance gains. This was a great article none the less. Thank You
“Is there really a need for anything that fast?” with 4 dual GPUs being used for CUDA/OPenCl application you do need even more than what is being provided to keep the GPUs filled with data with bottleneck.