NVELO Dataplex SSD Caching Software Review – Seven mSATA SSDs Prove An Amazing Concept

TEST BENCH AND SSD ROUNDUP

Our test bench is a Lenovo Thinkpad x220 Series laptop.  It is configured with an Intel Core 15-2520M CPU and 4GB RAM and running Windows 7 64bit OS.

The hard drive that we are going to be using for our test environment is the Lenovo standard Hitachi SATA 2 320GB hard drive.

As always, a quick click on any of our photos will give you a high resolution view.

SYSTEM DISASSEMBLY

The Lenovo X220 is an ideal candidate for testing of this type because, unlike all other manufacturers that I know of, videos and tutorials are publicly available that help you through system disassembly and the replacement of Thinkpad X220 hardware.

For our mSATA replacement, it was actually quite simple.  There are seven crews on the bottom that have pictures of a keyboard or trackpad beside them which we removed.  We then slid the keyboard towards the screen and lifted it.  From there, we pulled back on the one piece palm rest and track pad which afforded us total access to the mSATA card.  From there we just swapped each SSD as we went along.

BENCHMARK SOFTWARE

Our software selection for this review is a bit less detailed than most as we have decided to test all of the mSATA SSDs we could locate.  Software consists of ATTO Disk Benchmark, Crystal DiskMark, BootRacer and 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.

Benchmark software used by The SSD Review can be obtained by clicking on the title of each application as all may be downloaded without cost to the consumer.

HITACHI 320GB 7200RPM 3Gbps HARD DRIVE

The main purpose of this report is to validate that Dataplex can increase the performance of a hard drive close to that of an SSD.  Having said that, it would be unfair not to provide initial benchmarks for the hard drive itself.  Beware though because, as we were, you just may be truly stunned at the performance difference between the hard drive and hard drive cached.

ATTO DISK BENCHMARK VER. 2.46

ATTO Disk Benchmark is perhaps one of the oldest benchmarks going and is definitely the main staple for manufacturer specifications. ATTO uses RAW or compressible data and, for our benchmarks, we use a set length of 256mb and test both the read and write performance of various transfer sizes ranging from 0.5 to 8192kb. Manufacturers prefer this method of testing as it deals with raw (compressible) data rather than random (includes incompressible data) which, although more realistic, results in lower performance results.

ATTO results of 92MB/s read and 83MB/s write are typical of a hard drive and the hard drives access times are around 9ms, on average compared to 0.01ms for the SSD.  This is why a normal PC can take over a minute to boot compared to 15 seconds for the SSD system.

CRYSTAL DISK BENCHMARK VER. 3.0 X64

Crystal Disk Benchmark is used to measure read and write performance through sampling of raw (0/1 Fill/compressible) or random data which is, for the most part, incompressible.

Crystal DiskMark confirms the high sequential speeds of ATTO but take a look specifically at the 4k random write score of 0.803MB/s as this is the disk transfer method that is in use the most during typical operations and also responsible for the most visible user performance jump when we move to an SSD.

PCMARK VANTAGE X64 HDD SUITE

The SSD Review uses benchmark software called PCMark Vantage x64 HDD Suite to create testing scenarios that might be used in the typical user experience. There are eight tests in all and the tests performed record the speed of data movement in MB/s to which they are then given a numerical score after all of the tests are complete.

All eight tests are described in the results below and you might want to make a note of the fact that the highest transfer speed reached was 39MB/s with a Total Point Score of 4232.

25
Leave a Reply

avatar
9 Comment threads
16 Thread replies
0 Followers
 
Most reacted comment
Hottest comment thread
10 Comment authors
John LyuLes@TheSSDReviewMarco PennaMohdaliMRFS Recent comment authors
  Subscribe  
newest oldest most voted
Notify of
John
Guest
John

I have a question after some background information. I have an aging Dell Vostro 1500 notebook originally built to run Win Vista. It will allow use of a flash cache module (FCM) connectable using what appears to be a miniPCI connection. These miniSATAs appear to be miniPCI cards, but unlike the FCM that may have available drivers, I am unaware of drivers to make these miniSATAs work. If all the above is correct information, can I add one of these miniSATA’s cards to my Dell and have this caching software work with it to achieve this same goal. Or am… Read more »

Christopherkr
Guest
Christopherkr

I’m wondering the same thing as John. I have a latitude D630 which has the same FCM port, and I’ve not found out what it supports.

Anonymous
Guest
Anonymous

Are you really trying to tell me that a cached drive writes data faster than the SSD drive that is doing the caching???? This makes absolutely NO sense. Please explain. You can’t just throw out test results like that and then just expect people to accept the test’s veracity. What is the Dataplex software doing to the data that would cause this increase in write performance over a bare SSD? Is it some kind of compression? What am I missing? Also, I’m confused about bus speed. A SATA 3 SSD on a SATA 2 controller has to be slower than… Read more »

Les@TheSSDReview
Guest

The tests results are exactly as they are stated in the charts. As far as SATA 3 speeds on a SATA 2 controller, all SATA 3 SSDs are backwards compatible and we never found anything unusual in those results as well. We really don’t know what there would be to exlplain that might assist. All tests were done from the same laptop, all with the same HDD and caching software. The tests were simply the SSD scores and then the HDD cached scores from the specific SSD.

Anonymous
Guest
Anonymous

Am I being punk’d? Where is Ashton Kutcher? Did you even read my questions? What’s the difference whether the SSD is caching the data or just writing the data as a ‘normal’ SSD when it comes to write speed? Isn’t caching and writing the same thing? Obviously not, if the speed increases, but that then leaves the question regarding what is different unanswered. The only variable is the Dataplex software. So, I ask again…what is the Dataplex software doing that would increase write speed (higher than a standalone SSD)? How do you get BOTH increased access times AND increased sustained… Read more »

Les@TheSSDReview
Guest

I would like to help you but am afraid that there is really no way to do that. The results are the results and were consistent through a few tests, most commonly with the SF drives with respect to the increased write speed. Tx ahead.

Les@TheSSDReview
Guest

We have a bit of a return regarding the write speeds.

This is all in the noise. From the results, it looks like NVELO varies 5-10% better in some cases and 5-10% worse in other cases. With caching, the software write location/pattern is unique, such that sometimes it can outperform the SSD. In the future, it may be possible to do better more consistently based on the fact that the software recognizes that it is writing to SSD while the application thinks it’s just writing to some random storage device.

Anonymous
Guest
Anonymous

If this is the case, I would assume that as SSDs become more prevalent many software developers will begin to optimize their code for these drives.

Also, is there any way for The SSD Review to run tests to discern the difference in read/write speeds between a SATA 3 SSD running on a SATA 2 controller from a SATA 2 SSD running on the same SATA 2 controller. I’m trying to figure out if the controller or the drive is the main contributor to performance. Thanks.

Gerald Monroe
Guest

Fairly simple explanation : the writes are being performed by both the SSD and the HDD at the same time, each of them splitting the task respective to their relative speeds. Eventually the part of the data that was cached to the SSD will be moved to the hard drive, probably during either the next reboot or when the file system is not as busy.

NVELO is a software company with tens of employees and millions of dollars. It isn’t surprising that their product is more complex than you might expect.

Anonymous
Guest
Anonymous

Thanks for the info. Do you work for Nvelo? Could you point me to a website where this information is documented so I can read more about it? I’m curious how the software is able to write to two channels at the same time and how this ends up being faster than simply writing to the SSD. I didn’t know the bus speed was the weak link in the data storage process.

Les@TheSSDReview
Guest

I don’t work for NVelo and might suggest you contact them for additional information.

Lori
Guest
Lori

I got subscribed to an email list and don’t even know how…can’t find a way to unsubscribe and any attempts by email fell on blind eyes I guess

Craig
Guest

Les, I appreciate your team’s reviews, they are informative & I learn something useful every time I visit your site. Regarding the ability to cache data on a PC to increase performance: Any plans on reviewing other cache software other than Dataplex? Romex Software has a product in beta testing called FancyCache that uses System RAM as Level-1 cache & unseen RAM or an SSD as Level-2 cache. The performance is amazing. Unlike Dataplex, the Romex software is highly user adjustable & they have built in many safeguards to prevent system crashes & hang-ups (write flushing, etc). The Romex software… Read more »

Les@TheSSDReview
Guest

If you can put us in touch with this company or this company in touch with us, we would definitely be interested in taking a closer look.