Archive Home arrow Reviews: arrow Storage arrow ACARD ANS-9010 DDR2 SATA RAM-Drive
ACARD ANS-9010 DDR2 SATA RAM-Drive E-mail
Reviews - Featured Reviews: Storage
Written by Olin Coles   
Thursday, 26 March 2009
Table of Contents: Page Index
ACARD ANS-9010 DDR2 SATA RAM-Drive
Features and Specifications
First Look: ACARD ANS-9010
ACARD ANS-9010 Details
SSD Testing Methodology
Random Access Time Benchmark
Basic IOPS Performance
Linear Bandwidth Speed
Buffered Transaction Speed
Windows XP Startup Times
Solid State Drive Final Thoughts
ACARD ANS-9010 Conclusion

ATTO Disk Benchmark Results

EDITORS NOTE: ATTO Disk Benchmark is not designed to be used as bandwidth speed tool, as the final results are determined by user-set variables. Benchmark Reviews uses ATTO Disk Benchmark as a tool for illustrating basic IOPS load performance at various chunk load sizes. Please read the Solid State Drive Benchmark Performance Testing article to understand how the benchmarks used in this article should be interpreted.

The ATTO Disk Benchmark program is free, and offers a comprehensive set of test variables to work with. In terms of disk performance, it measures interface transfer rates at various intervals for a user-specified length and then reports read and write speeds for these spot-tests. There are some minor improvements made to the 2.34 version of the program, but the benchmark is still limited to non-linear samples up to 256MB. While the bandwidth results are no realistic for determining the maximum drive speeds, ATTO Disk Benchmark is still a good tool for illustrating the bandwidth at different file size chunks. Please consider the results displayed by this benchmark to be basic IOPS performance indicators.

Beginning with the integrated Intel ICH10R Southbridge chip connected to the ACARD ANS-9010, the ATTO Disk Benchmark tools performs file transfers ranging from 0.5 KB to 8192 KB. The DDR2 RAM-Disk shows a read and write plateau ranging from 128-4096 KB file chunks. Because the ACARD ANS-9010 can be configured so many different ways (depending on the number of DIMMs populated and density of modules), Benchmark Reviews has decided to offer perspective on several various set-ups.

ACARD_ANS-9010_ATTO_2x-DIMM.png ACARD ANS-9010 4GB @ 2 DIMMs

ACARD_ANS-9010_ATTO_4x-DIMM.png ACARD ANS-9010 8GB @ 4 DIMMs

ACARD_ANS-9010_ATTO_2x-DIMM_RAID-0.png ACARD ANS-9010 4GB/2 DIMMs in RAID-0

ACARD_ANS-9010_ATTO_4x-DIMM_RAID-0.png ACARD ANS-9010 8GB/4 DIMMs in RAID-0

After watching the peak read bandwidth measure 187 MBps and write-to speeds max at 153 MBps, I was totally shocked by the RAID-0 performance. On its own merit, the standard (P0) single SATA connection did quite well, and 187/153 isn't at all bad. However, I don't expect anyone to be using the ANS-9010 as a primary drive. This is a RAM-Disk, and most people who use these products plan to utilize it more for speed; RAM-Disks are not typically used for data storage.

So RAID-0 is clearly what ACARD intended for their ANS-9010 RAM-Disk, and that's exactly how Benchmark Reviews will test it. Looking over the various results for the RAID-0 configuration, you can see that the using only two DDR2 modules to populate the inner (yellow) DIMM slots offer the best speeds. With only two DIMMs occupied with 4GB or DDR2 RAM, the ACARD ANS-9010 performed at 510 MBps read and 157 MBps write. The read-from bandwidth was phenomenal, and we've yet to see another product come anywhere near that speed. The write-to bandwidth wasn't so impressive, as it remained nearly identical to the non-RAID configuration.

Drive Hardware

In our next section, we test linear read and write bandwidth performance of the ACARD ANS-9010 RAM-Disk and compare its speed against several other top storage products. Benchmark Reviews feels that linear tests are the best method for testing SSD products, as detailed in our Solid State Drive Benchmark Performance Testing article.



 

Comments 

 
# RE: ACARD ANS-9010 DDR2 SATA RAM-DriveJeddo 2010-03-12 16:11
Why bother with a hardware device that uses RAM modules at such low speeds? I mean, 500 MB/s may be fast for anything plugged into a SATA socket, but with modern consumer motherboards accomodating 16 GB RAM, take half of that for a RAMdrive and you'll be far better off in terms of raw performance. I measured approx. 2 GB/s on a 1 GB RAMdrive; system: Athlon 64 x2 6000+, 2 GB DDR2, XP SP3. And if 4 RAM sockets are not enough, I'm sure Tyan can meet your needs.
Report Comment
 
 
# RE: RE: ACARD ANS-9010 DDR2 SATA RAM-DriveDennis 2010-06-12 19:18
There is a definite niche market for this product. If you have a 32-bit OS like WinXP (and are forced to use this), and run a memory hungry app like PhotoShop, you will need something to help you through the 4GB memory barrier. This handy drive will take care of all you paging file, virtual memory, and temporary drive storage needs. It will provide a strong boost in performance. I'm planning on getting one shortly for use in a corporate environment currently locked to Windows XP 32-bit.
Report Comment
 
 
# Access speed is more important for SSDs.donjoe 2011-01-21 05:31
"the average linear read and write bandwidth [...] I personally consider this the single most important comparison of storage drive products"

You personally have no idea what you're talking about - SSDs are way too expensive per GB to be used mainly for long-term storage, so it makes no sense to talk about them as being anything other than active, system-and-applications or server drives, where the ACCESS SPEED is the key factor, not the linear I/O speed.
Report Comment
 
 
# RE: Access speed is more important for SSDs.Olin Coles 2011-01-21 08:27
How do you take that quote to mean that I endorse SSDs for primary storage? Considering how many SSDs I've personally tested (over fifty), and how many more I've personally deployed on servers and workstations (thirty), I think you might want to reconsider your baseless claim.

What I'm saying here is that linear I/O is more important than the other factors, since SSDs will already inherently deliver faster access speed and transfer bandwidth. In my opinion, measuring input/output operations is more important than measuring speed or response time on an SSD.
Report Comment
 
 
# RE: ACARD ANS-9010 DDR2 SATA RAM-Drivethrakkor 2011-12-28 11:15
I suspect that the main limitation here is the bridge chip. The RAM, even DDR2, has huge speed headroom (proven by mainboard chipsets); the SATA interface may or may not stand in the way of really high transfer speeds, but given that many modern days Flash based SSDs have competitive speeds, the remaining factor is the bridge chip, along with the design of the board. I really expect this kind of device, the RAM based SSD to offer over 300 megabytes / second read and write, easily, no RAID required.
Conclusion: SODIMM DDR3 for improved RAM density (16 slots would be really top), SATA3 interfacing (there aren't even may SATA3 HDDs) and an improved bridging design (even a small dedicated computer of its own would be necessary to handle the huge bandwidth). Since there are many 300 to 400 USD graphic monsters that are used just for games - what a stupid purpose - I don't even blink at spending, say, 500 USD on such a future DDR3 based SSD, and that's before any SODIMM would make its way into the slots. I don't even blink at that price. But, please, 150 MB/s ? Hah... More like 300 - 400, SATA3 or even SAS...
Report Comment
 
 
# RAMhirnaxi karelia 2012-06-12 14:27
hi i want to know about how can i find ram; speed, pins, title, and eles.
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