ACard’s ANS-9010 Serial ATA RAM disk

Manufacturer ACard
Model ANS-9010
Price (Street)
Availability Now

Solid-state storage is the new hotness. Of course, that hotness is relative—this is still the storage market we’re talking about. Think less Scarlett Johansson and more Tina Fey. Nonetheless, the recent almost-affordability of blazing-fast flash drives has been an interesting development in the storage world. And it’s getting even more interesting, because in addition to an entire family of flash drives, the SSD market also has a crazy uncle.

The focus of most solid-state storage discussions centers on flash-based solutions, but they’re not the only game in town. DRAM chips offer solid-state storage, too, and they’re widely available on plain old memory modules. What’s more, even Intel’s fastest X25-E flash drive is limited to 250MB/s reads and 170MB/s writes, but your average DDR2-800 DIMM can read and write at up to 6.4GB/s. That’s like lining up a 40-horsepower motor ripped from a Vespa against the W16 monstrosity that powers the Bugatti Veyron.

Of course, you still need a way to get DRAM chips to behave as a hard drive. The hardware to do so was initially offered only for high-end servers and workstations, and it was priced accordingly. Then, about three years ago, Gigabyte went out on a limb with an i-RAM storage device that allowed users to plug four DIMMs into a PCI card that hooked up to a standard Serial ATA port. The i-RAM was ridiculously fast, and for most intents and purposes, it behaved exactly like a normal hard drive.

The original i-RAM’s total capacity was limited to 4GB of DDR SDRAM, so it’s not particularly useful by today’s standards. However, a company called ACard has done one better with the ANS-9010 RAM disk, which has eight DDR2 DIMM slots and support for up to 8GB of memory per slot. The ANS-9010 also features a pair of Serial ATA ports, allowing it to function as a single drive or masquerade as a pair of drives that can easily be split into an even faster RAID 0 array. Can this latest RAM disk live up to its staggering performance potential, and perhaps more importantly, could you live with it as a primary hard drive?

The difference DRAM makes

Before taking a closer look at the ANS-9010, I should date myself by noting that the RAM disk concept is really nothing new. Those who have been around for a while will remember that sectioning off a chunk of system memory for storage was all the rage back in the day. RAM disks were a software solution back then, so they used memory already plugged into your motherboard. The move to hardware-based RAM disks has made things much easier by allowing users to access memory modules through a standard hard drive interface without the need for drivers or even an operating system.

So why use DRAM over the flash memory everyone and their mother seems to have their fingers in? Speed, for one. I’ve already mentioned DDR2’s 6.4GB/s transfer rates. However, that 6.4GB/s is just a theoretical peak. The Serial ATA interface is currently limited to 300MB/s, and even with both its SATA ports active, ACard only claims transfer rates up to 400MB/s for the ANS-9010.

Another benefit DRAM has over flash memory is that there’s no limit on the number of write-erase cycles it can endure. Effective wear leveling algorithms and single-level cell memory can greatly improve the lifespan of a flash drive, but they just prolong the inevitable. DRAM’s resiliency does come with a cost, though. While flash memory cells retain their data when the power is cut, DRAM is volatile, so it does not. To keep DRAM data intact, you have to keep the chips juiced.

Evolving the RAM disk

Storage solutions generally aren’t much to look at, but the ANS-9010 is a little different. Instead of being tucked away in an internal hard drive bay, it slides into an external 5.25″ optical bay where you can see it.

The device’s front panel has a few ventilation slits and is littered with indicator lights for power, drive access, and battery life. ACard equips the ANS-9010 with a 7.4V, 2400 mAh battery that keeps the drive’s memory powered when the host system is shut off. Our battery was able to keep 16GB of memory powered for more than four hours, which isn’t nearly good enough if you want to turn your system off at night, but more than enough to get through the occasional reboot or hardware upgrade.

For those who would rather not trust a battery with preserving their data, the ANS-9010 also features a CF card slot up front. With the push of a button, the drive can back up or restore the contents of its memory to a flash card. This backup capability might seem like a small thing, but it greatly improves the RAM disk’s usability in the real world. ACard quotes a CF backup time of 21 minutes and 34 seconds for a 32GB configuration and a restore time of just over 14 minutes for the same config.

Busting the ANS-9010 open reveals eight 240-pin DDR2 DIMM slots. ACard says module densities up to 8GB are supported, bringing the drive’s total capacity potential up to an impressive 64GB. Keep in mind that 8GB memory modules are extremely expensive, though. For most folks, cost alone will limit the drive’s total capacity to 32GB or less. ACard doesn’t yet list any 8GB modules on its memory compatibility list, either. You’ll want to pay attention to that compatibility list, because registered and ECC DIMMs aren’t supported. It also took a couple of firmware revisions for the ANS-9010 to get along with the OCZ DIMMs we used for testing.

The RAM disk’s DIMM slots are arranged in pairs of four, with spacing similar to what you’d find on a motherboard. This leaves just enough room for modules with normal heatspreaders, although taller DIMMs may bump their heads on the 5.25″ drive bay ceiling. You can still run taller DIMMs by removing the ANS-9010’s top panel, and as long as you don’t have a 5.25″ drive in the bay directly above, most cases should be accommodating.

Between the rows of DIMM slots sits the RAM disk’s battery and really the only silicon of note. Hidden under a passive heatsink sits a FPGA chip that ACard has programmed to make the memory modules behave as a hard drive. As Intel’s X25-series SSDs have illustrated, a good storage controller architecture can significantly improve performance. ACard isn’t saying much about the ANS-9010’s storage controller design, but we’ll soon see whether it’s made the most of the prodigious bandwidth potential inherent to DDR2 memory.

We do know that ACard’s storage and memory controller has at least one interesting trick up its sleeve—it’s able to split installed memory between two physical Serial ATA ports. Even with only a single DIMM installed, the ANS-9010 can present itself as two physical hard drives via two SATA ports located at the rear. This feature allows users to wring even more performance from the drive by pairing it with a RAID controller for a little striped RAID 0 love. And because solid-state storage is immune to the mechanical failures that make running RAID 0 dicey with traditional hard drives, you don’t have to worry about a head crash hosing the array.

Switching the ANS-9010 between single-drive and RAID modes requires flipping a jumper, which is accessible from the rear. Otherwise, there isn’t much to see. The drive pulls juice from a standard Serial ATA power cable, although it only seems to draw from the 12V line.

Bring on the DRAM

The ANS-9010 arrives empty, so you have to provide your own memory. Today we’ll be testing the drive loaded with 16GB of DDR2-800 memory courtesy of OCZ. DDR2-800 is cheap and plentiful these days, and it’s often less expensive than slower speed grades that should still provide more than enough bandwidth for a RAM disk. These Gold Series modules can be found in two-DIMM, 4GB kits for as little as $44 online, pegging the cost of 16GB at well under $200. Thanks to OCZ for providing these modules for testing.

Our testing methods

We’ve assembled motley crew of competitors to line up against the ANS-9010 today. In addition to the fastest mechanical desktop drives from Hitachi, Samsung, Seagate, and Western Digital, we’ve thrown in a few flash-based SSDs from Intel and Samsung. And since the i-RAM started this whole business of inexpensive hardware RAM disks, we’ve included it, as well.

To keep the graphs on the following pages easier to read, we’ve color-coded the bars by manufacturer. The ANS-9010 appears in bright red, and we’ve tested it in single-drive and RAID 0 configs. In RAID mode, we relied on the RAID feature built into our test system’s Intel south bridge.

All tests were run three times, and their results were averaged, using the following test system.

Processor Pentium 4 Extreme Edition 3.4GHz
System bus 800MHz (200MHz quad-pumped)
Motherboard Asus P5WD2 Premium
Bios revision 0422
North bridge Intel 955X MCH
South bridge Intel ICH7R
Chipset drivers Chipset 7.2.1.1003
AHCI/RAID 5.1.0.1022
Memory size 1GB (2 DIMMs)
Memory type Micron DDR2 SDRAM at 533MHz
CAS latency (CL) 3
RAS to CAS delay (tRCD) 3
RAS precharge (tRP) 3
Cycle time (tRAS) 8
Audio codec ALC882D
Graphics Radeon X700 Pro 256MB with CATALYST 5.7 drivers
Hard drives Seagate Barracuda 7200.11 1TB
Seagate Barracuda ES.2 1TB
Samsung SpinPoint F1 1TB

Hitachi Deskstar E7K1000 1TB

Western Digital VelociRaptor 300GB

Western Digital Raptor WD1500ADFD 150GB


Western Digital Caviar Black 1TB


Western Digital RE3 1TB


Western Digital Caviar SE16 640GB



Seagate Barracuda 7200.11 1.5TB


Samsung FlashSSD 64GB


Intel X25-M 80GB


Intel X25-E Extreme 32GB


Gigabyte i-RAM
with 4GB DDR400 SDRAM

ACard ANS-9010
with 16GB DDR2-800 SDRAM

OS Windows XP Professional
OS updates Service Pack 2

Thanks to NCIX for getting us the Deskstar 7K1000 and SpinPoint F1.

Our test system was powered by an OCZ PowerStream power supply unit.

We used the following versions of our test applications:

The test systems’ Windows desktop was set at 1280×1024 in 32-bit color at an 85Hz screen refresh rate. Vertical refresh sync (vsync) was disabled for all tests.

All the tests and methods we employed are publicly available and reproducible. If you have questions about our methods, hit our forums to talk with us about them.

WorldBench
WorldBench uses scripting to step through a series of tasks in common Windows applications. It then produces an overall score. WorldBench also spits out individual results for its component application tests, allowing us to compare performance in each. We’ll look at the overall score, and then we’ll show individual application results. You won’t find Gigabyte’s i-RAM in the graphs below because its 4GB maximum storage capacity is too limited for WorldBench to run.

The ANS-9010 gets off to a good start in WorldBench, besting the X25-E Extreme by a single point. Running the drive in RAID mode bumps its overall score by two points, which is less than one might expect. However, keep in mind that WorldBench runs through common desktop applications that tend not to be bound by storage subsystem performance. Let’s step through the individual application results to see how the ACard fares in each test.

Multimedia editing and encoding

MusicMatch Jukebox

Windows Media Encoder

Adobe Premiere

VideoWave Movie Creator

Although it stays at the front of the pack through most of WorldBench’s multimedia editing and encoding tests, the ANS-9010 is a hair behind Intel’s fastest SSD in the one benchmark that spreads the field. In the Premiere test, the X25-E comes out ahead of the ANS-9010 by one to four seconds depending on whether the RAM disk is running in single-drive or RAID mode.

Image processing

Adobe Photoshop

ACDSee PowerPack

Another member of the WorldBench suite that benefits from faster storage is the ACDSee test, which has the ANS-9010 making an SSD sandwich out of the X25-E.

Multitasking and office applications

Microsoft Office

Mozilla

Mozilla and Windows Media Encoder

The field doesn’t spread much in WorldBench’s office and multitasking tests, although the ANS-9010 continues to turn in some of the quickest completion times of the lot.

Other applications

WinZip

Nero

Among all of WorldBench’s individual application tests, Nero and WinZip are the most demanding of the storage subsystem. In the latter, which appears primarily to stress read performance, the ANS-9010 just edges out Intel’s SSDs. The RAM disk stretches its legs even more in Nero, with our single-drive config outpacing the X25-E by 13 seconds. The RAID 0 setup is an additional 20 seconds faster than that. This test appears to have a mix of read and write requests, as evidenced by the performance gap between the X25-E and the X25-M; the X-25M’s write speeds are 100MB/s slower than those of the X25-E Extreme.

Boot and load times
To test system boot and game level load times, we busted out our trusty stopwatch.

The i-RAM makes its first appearance in our boot time test, and it’s roughly as fast as ACard’s RAM disk. However, both are lodged in the middle of the pack and much slower than not only Samsung’s FlashSSD, but also a few mechanical hard drives.

The RAM disks claw back to the front of the pack in our level load tests. The i-RAM and ANS-9010 share the lead in Doom 3, but they’re not quite as quick as the X25-M and FlashSSD in Far Cry.

File Copy Test
File Copy Test is a pseudo-real-world benchmark that times how long it takes to create, read, and copy files in various test patterns. File copying is tested twice: once with the source and target on the same partition, and once with the target on a separate partition. Scores are presented in MB/s.

To make things easier to read, we’ve separated our FC-Test results into individual graphs for each test pattern. We’ll tackle file creation performance first.

The ANS-9010 easily outguns the competition in FC-Test’s file creation tests. Only the i-RAM and X25-E Extreme give it much competition here, and they’re really only within striking distance of the single-drive config in two of the five test patterns. Striping makes the ANS-9010 even faster, although it’s hardly the doubling of performance one might expect—or at least hope for—from such an arrangement.

Interestingly, the ANS-9010 stumbles a little when FC-Test moves from writes to reads. ACard’s RAM disk is by no means slow, but it’s not as quick as Intel’s X25-series SSDs, which take the first two spots through four of five test patterns. Only with the Windows test pattern, which is made up of a large number of small files, does the ANS-9010 eke out a win, and then only when running in RAID 0.

FC-Test – continued

Next, File Copy Test combines read and write tasks in some, er, copy tests.

The ANS-9010 returns to glory in FC-Test’s suite of copy tests, although it takes RAID to outrun the X25-E Extreme with the ISO test pattern, which is made up of a small number of very large files. Note that the gap between the RAM disk’s single-drive and RAID configs is significant with some test patterns. The ANS-9010 is also notably faster than the i-RAM across the board.

Next, we turn our attention to FC-Test’s partition copy tests, which copy data between partitions on each drive. The i-RAM’s 4GB maximum capacity precludes it from running these tests, so you won’t find Gigabyte’s RAM disk listed in the graphs below.

The partition copy tests shake out much like the first wave of copy tests, except this time the single-drive ANS-9010 sweeps the X25-E Extreme across all five test patterns. Keep in mind that the Extreme is Intel’s most expensive SLC-based solid-state drive. The X25-M uses less expensive MLC flash memory, whose comparatively slow write speeds really hurt the drive’s performance here.

iPEAK multitasking
We’ve developed a series of disk-intensive multitasking tests to highlight the impact of seek times and command queuing on hard drive performance. You can get the low-down on these iPEAK-based tests here. The mean service time of each drive is reported in milliseconds, with lower values representing better performance.

Our iPEAK workloads were recorded using a 40GB partition, so they’re a little big for the 4GB i-RAM, 16GB ANS-9010, and even the 32GB X25-E. The app had no problems running, but it warned us that I/O requests that referenced areas beyond the drives’ respective capacities would be wrapped around to the beginning of each drive. Since there should be no performance difference between the beginning and end of an SSD, the results should be valid.

When configured in a RAID 0 array, the ANS-9010 runs the table in iPEAK, turning in the quickest mean service times across all nine workloads. This RAID setup is notably quicker than running the ANS-9010 in single-drive mode, too. If we average the response time of each config across all workloads, ACard’s RAM disk turns in a 0.18-millisecond service time in RAID and a 0.32-millisecond service time in single-drive mode, making it nearly twice as fast when you use both SATA ports.

Of course, the single-drive config’s slower performance leaves it vulnerable to the X25-E in some workloads. In four of our multitasking tests, the Extreme just edges out the single-drive ANS-9010. ACard’s take on the RAM disk concept is consistently faster than Gigabyte’s four-year-old i-RAM, though.

IOMeter
IOMeter presents a good test case for both seek times and command queuing.

Wow. That’s, uh. Wow. With the exception of the web server test pattern, which is made up entirely of read requests, the ANS-9010 completely dominates our IOMeter transaction rate tests. Even in single-drive mode, it’s quicker than the i-RAM and much faster than the X25-E. The Intel SSDs prove worthy competition with the web server test pattern, but they’re still trumped by the ANS-9010 running in RAID mode. Splitting ACard’s RAM disk into a striped array just about doubles its transaction rates at higher load levels.

Of course, there’s a price to be paid for those high transaction rates. The ANS-9010 also consumes quite a lot of CPU power in the process. To put these results into perspective, we’ve whipped up another set of graphs that illustrates the transaction rate per percent CPU utilization. Since our mechanical hard drives don’t deliver anywhere near SSD levels of performance here, we’ve left them out of the equation, with the exception of the VelociRaptor.

The ANS-9010’s RAID config may be exceptionally fast, but there’s a greater CPU utilization cost for each transaction than with a single-drive setup. Interestingly, the i-RAM and the single-drive ANS-9010 are evenly matched here. However, Intel’s SSDs have the highest transaction rates per CPU cycle overall.

HD Tach
We tested HD Tach with the benchmark’s full variable zone size setting.

So much for DRAM’s peak bandwidth advantage over flash memory. In HD Tach’s sequential transfer rate drag race, the ANS-9010 only manages to read at 173.6MB/s and write at 144.1MB/s. What’s more, it’s slightly slower when running in RAID 0. Those sustained transfer rates do put ACard’s RAM disk ahead of the i-RAM, but the X25-E Extreme is much faster. Even the X25-M is quicker in the read speed test.

The ANS-9010’s burst performance isn’t as quick as one might hope, either. The drive doesn’t even manage 200MB/s with one SATA port plugged in, although it does eclipse 300MB/s when running in RAID.

Solid-state drives do exceptionally well in random access time tests, and the ANS-9010 is no exception. The drive’s seek times are essentially instantaneous, and interestingly, they’re just a little bit quicker when running in single-drive mode. I’m hesitant to make too much of the 0.1-millisecond difference between single-drive and RAID modes, but it was consistent across three test runs, suggesting that there’s a hint of a latency penalty associated with splitting the RAM disk’s memory between two SATA ports.

HD Tach’s margin for error in the CPU utilization test is +/- 2%, which narrows our results considerably.

Noise levels
Noise levels were measured with an Extech 407727 Digital Sound Level meter 1″ from the side of the drives at idle and under an HD Tach seek load. Drives were run with the PCB facing up.

Because it has no moving parts, the ANS-9010 is completely silent. The 42.6-decibel scores you see for the SSDs represent the ambient noise levels of the test system itself.

Power consumption
For our power consumption tests, we measured the voltage drop across a 0.1-ohm resistor placed in line with the 5V and 12V lines connected to each drive. Through the magic of Ohm’s Law, we were able to calculate the power draw from each voltage rail and add them together for the total power draw of the drive. Because the i-RAM pulls its power from a PCI slot, we were unable to measure its exact power draw and include it in the results below.

Regardless of what it’s doing, the ANS-9010 pulls just under 12 watts with 16GB of memory spread across all eight of its DIMM slots. The drive’s power consumption is significantly higher than that of flash-based SSDs and even most of our hard drives, but it’s not entirely unreasonable.

Conclusions

I have to admire ACard for coming up with the ANS-9010 RAM disk. Not only does it make the solid-state storage market a little more interesting, but it occasionally offers truly inspiring performance, as well. The drive easily handled our disk-intensive iPEAK multitasking and multi-user IOMeter loads, besting the fastest SSD on the market not only when running in RAID mode, but more often than not, in a single-drive config, too. However, the ANS-9010’s dominance wasn’t universal. Sure, it was the quickest drive of the lot in WorldBench, but not by a significant margin. And although it exhibited the fastest write and copy speeds in FC-Test, Intel’s SSDs proved quicker when it came time to reads.

The battle between the ANS-9010 and Intel’s X25-E Extreme SSD was certainly an intriguing one to watch, in particular because the Extreme’s storage controller seems to be the best in the business. A relatively poor performance in HD Tach’s sustained and burst transfer rate tests suggests this latest RAM disk still has room to improve on that front. Sure, it handily beat the i-RAM across the board, but Gigabyte’s take on the inexpensive RAM disk concept is several years old.

Despite leaving some DRAM performance potential on the table, ACard has done a number of smart little things to make the ANS-9010 easier to live with and more attractive than the original i-RAM. With eight DIMM slots and support for configurations up to 64GB, there’s plenty of storage capacity to be had. The CF backup feature is really nice given DRAM’s volatility, too. And let’s not forget the ANS-9010’s hint of memory mitosis, which splits available capacity for RAID configurations that can, in some cases, almost double performance.

So how much does all of this cost? $380 for the ANS-9010, plus the cost of memory. The 16GB of DDR2-800 DIMMs we used today will set you back roughly $176, bringing the drive’s total cost up to about $555. A 32GB configuration made up of 4GB modules, which run around $100 each, will set you back just under $1200 when all is said and done. This latest RAM disk isn’t cheap, then. But neither are SSDs that offer equivalent performance. Intel’s 32GB X25-E Extreme runs around $600, for example.

For most applications, the fact that the X25-E Extreme comes in a smaller form factor, consumes much less power, and actually has a lower cost per gigabyte than the ANS-9010 makes the RAM disk a tough sell. However, the ANS-9010 can be much faster under the right circumstances, and there’s less reason to worry about the drive eventually wearing out.

Even if it may ultimately be relegated to a niche of the ultra-high-performance storage market, I quite like ANS-9010. ACard should be applauded for doing something a little different. And speaking of a little different, you can also get an ANS-9010B RAM disk for about $250. This B revision is limited to six DIMM slots and 24GB of memory, and it lacks the second SATA port for RAID, but it’s nice to have the option of a cheaper alternative with fewer frills.

Comments closed
    • blubje
    • 10 years ago

    I just completed a review of this device with many more tests showing the performance in 1.5gb SATA mode and 3.0 mode.
    Check it out -> §[< http://2xod.com/articles/ANS_9010_ramdisk_review/<]§

    • WareWolf801
    • 11 years ago

    Excellent article. I’ve been waiting for something like this for a long time, with testing and all, nice job really. I’ve wanted for a product like this for years. The iRam was close, but limited & required special drivers..meaning it couldn’t work with freaky os’s sometimes (centOS for example).

    I recently bought a SSD which uses an inexpensive MLC technology I think (maybe SLC, if I’m lucky). It’s called a filemate goSolid .. interesting name thats for sure. Anyways, it’s 64gb, and it hauls butt 100+/- mb/sec read and 80+/- write! Additionally, I have 2 of the Seagate Baracuda’s you tested! Wow, cool! Likewise, seeing my friends’ raptors in there was really helpful. I’ve heard of and seen the F1 drive around, thanks for including it. Also, thanks for including ALL these sweet products in the testing..some of us DO realize that buying and using better components can pay off big time & applies to storage as much as the memory you use in a system, or its processor.

    About the product: This is a unique product for a certain class of customer, especially given the premium price. The other models mentioned from same mfg. might be a little more usable by the mainstream, and could really be the sweet-spot that sells millions of units.. Very cool technologies are being developed here with this unit, it could really be the shiz for use in a fileserver.. I wonder what a storage array would do with a gang of these. I wonder what a [DRAM!] cached RAID controller could do with a gang of these monsters? Any XDDR (like in my ps3?) RAID cards out there yet? What about a ddr3 solution from the same co!? Gawd man, 600MB/sec writes could possibly happen then?! FFS!

    The only thing I would say you could have added to the article was a little more about compatibility, did it work in linux, if so, which types? Since you tested transactions: Will it run with informix, or a postgres/mysql/oracle db-testing suite against this might be interesting? What about webserver support: centOS, debian, wha? Can I plug this thing into my psp and play games off iso files thru a usb to sata2 connector? Just kidding man, you did a great article, thanks I appreciate it.

    • Trymor
    • 11 years ago

    If I’m not mistaken, the older versions of the hyperdrive predated the original i-RAM.

    That being said, I would buy (at least) one of these if the prices were $100 for the low end model, and $150 (U.S.) for the high end model, as I have been waiting for an i-RAM type drive in any form with a SATA-2 interface, AND a resonable price. Having now watched some youtube videos on these DRAM hard drives booting and using various OS’s, I want one even more.

    Unfortunately, the pricing is too high for the timing of the re-release of this product.

    EDIT: It would be interesting to see TR review the ACARD ANS-9012 2.5 inch SATA-to-SDHC Flash Disk. I can think of a few uses for that relativly cheap product with some cheap fast SD cards in it….

    • issa2000
    • 11 years ago

    say hyperdrive 5

    • issa2000
    • 11 years ago

    this is old news

    it been around chine product made for

    hyperdrive 5 (say hyperdrive 4)

    re-released under new name:

    • Kurlon
    • 11 years ago

    Use for an I-RAM or other ram based SSD: Windows 7 on an eMachines eTower 633i

    I’m using it as my only swap volume at the moment, initially dedicated it to Readyboost but this seems to work better. No Aero Glass as the last i810 graphics drivers ever tossed out were for XP, but other than that, it works damn well.

    It’s feels faster on the desktop/browsing/etc than it did with Fedora 10 or Debian “Lenny” on it. I did have to find a tweaked bios originally destined for an HP to install though, as the ACPI implementation is woefully antique.

    So far I’ve got an SATA controller shoved in, Gigabyte I-RAM with 4GB for swap, 512MB SDRAM (max it’ll take), and a P3 1ghz running at 750mhz thanks to the lack of 133mhz FSB support. If I had a third usable PCI slot I’d find another I-RAM to shove in so one could swap and one could handle ReadyBoost… got a 500GB IDE drive to throw at it, and I may try to track down a 1.4ghz Tualatin Celeron and at that point it’ll be ‘maxxed out’.

    So, if anyone wants to suggest benches to test the impact of RAM drives on slow boxes, or ReadyBoost, I’ve got the setup for it now. : )

    • Kurlon
    • 11 years ago

    There is an external power supply available. And I believe the cost is due to the use of an FPGA rather than custom silicon. (That’s likely a good chunk of the bottleneck as well.)

    For all the people clamoring for a PCIe version, that requires drivers to do any good. Going SATA means this can drop into a whole ton of systems as is. It’s a great way to speed up otherwise hammered systems that can’t take any more ram or are otherwise disk bound without having to re-invent the wheel.

      • Anonymous Coward
      • 11 years ago

      Yeah custom drivers would be fine. It wouldn’t be terribly hard to write a driver for what is essentially RAM (once the frivolous and expense adding crap like dump-to-flash is removed). I would love PCIe “file cache cards” … drivers and all.

    • erikejw2
    • 11 years ago

    Great review.
    I really like these a little more exotic hardware solutions.
    This kind of technology has great promise.

    Unfortunately the this product fail miserably in several aspects.

    1. 400MB/s throuhgput, come on, use a 4x pci-express and get some decent number instead.

    2. 4 hours battery life, ridiculous, add an external power adapter, you can get one from China for 1 dollar

    3. price, 380$. I can see myself by this for 150$ but 380 and they price themselves out of the market. I’d much rather buy 2 32 GB SSDs and put them in RAID0. That makes much more sense than this. SSD prices will go down very quickly this and the coming years. I guess this card cannot cost much more than 30$-40$ to produce, this premium price will only hurt themselves in the long run.

    ———–

    I can live with 16Gb(8x2Gb) of boot drive if I get an extreme performance boost compared to SSDs. In the coming months there will be 500Mb/s+ SSDs that we can RAID. We need something like 2GB/s+ for this card to be interesting.

    Interesting concept and a good try but go back and make a new product with some small changes and I will buy it.

    • marvelous
    • 11 years ago

    I was always interested in Ram drive. Those intel x25 are pretty fast. I would have never thought it would be fast as a ddr2 ram drive.

    I would have liked if Geoff included ram drive directly from our motherboard and see how well this contraption compares.

    • FireGryphon
    • 11 years ago

    Grammar error, page 5: “The RAM disks *[

      • Damage
      • 11 years ago

      Strange. Looks like a typo to me. 😉

    • mattthemuppet
    • 11 years ago

    shouldn’t that be meiosis – mitosis would be RAID 1 (identical copies)?

      • Dissonance
      • 11 years ago

      I’m referring to the splitting, not necessarily the nature of the offspring. You can put the pseudo-two-drive RAM disk in a RAID 0 or 1 array.

    • Bion1c
    • 11 years ago

    my first thought was “i’ll take 2 and run 4-way raid0 off my hardware raid controller”

    .. but then i saw the price 🙁 it’s too high for the numbers it puts out

    I don’t want to hear complaints about “expensive raid controllers” people.. anyone with a half decent storage setup these days probably already has a few pci-e raid cards lying around at this point, and if you dont, you can buy a perc 5/i off ebay for ~USD$120

    Besides, what we really want to know is the true performance of the device being tested – why bottleneck it with a crappy raid controller?? maybe we should run it in SATA1 mode too?

    • swaaye
    • 11 years ago

    Some of those benches sure do show that storage IO isn’t the bottleneck all that often.

    • ssidbroadcast
    • 11 years ago

    The front of this thing will make your computer look like Kit from Night Rider. Good or bad thing? That’s up to you.

      • WareWolf801
      • 11 years ago

      Those are just cooling vents man..

      As for Knight-rider lights are hard to find, but possible: dragonmodz.net..hdd activity light + they have a light-bar which could be re-wired to light in series like K.I.T.T.

    • SpotTheCat
    • 11 years ago

    The age of flash drives is FAST approaching. I think this kind of thing is past it’s prime release date.

    I don’t mind the whole failure after however-many writes thing of flash memory. They’ll still last beyond their useful life, and anybody using mechanical hard drives as long-term backup solutions are kidding themselves. Sure I have an odd 20GB hard drive from 8 years ago or so, but I took everything important off of it years ago.

      • Krogoth
      • 11 years ago

      HDDs are still going to around for a long, long time.

      Flash drives are not a complete replacement as they cannot match magnetic media on a GB/$$$$ ratio. Flash Drives are just going render the fast RPM, low-capacity HDDs obsolete.

        • SpotTheCat
        • 11 years ago

        You don’t understand exponential growth, do you? Mark my words, flash drives will replace mechanical disks in 5 years for single-user systems in all but the highest storage densities.

          • MadManOriginal
          • 11 years ago

          You’re right for a number of reasons. First is magnetic media is reaching limits of density, we’ve seen an explosion in disk size in absolute terms in the last few years because of perpendicular recording which had been in the works for a long time but what’s after that for magnetic media? It will be incremental increases rather than huge leaps.

          And then there’s flash density/$ increase rates. For the last 2-4 years flash has doubled in density and halved in price, for a total $/byte drop of 1/4 per year. With simple silicon like flash the barriers of incremental improvements are further off than for magnetic media. in one year I’m thinking we’ll see universally good-performing SSDs at under $1/GB…5 years? Unless there’s another breakthrough like PMR for magnetic media we’ll see SSDs become so cheap that magnetic will go by the wayside.

            • Krogoth
            • 11 years ago

            You pretend like Flash will never hit limitations, when it could very well happen to it as well.

            Will flash get better in $$$$/GB and performance ratio? Yes

            Will it ever match magnetic media in $$$$/GB ratio? Not likely to happen anytime soon.

            • MadManOriginal
            • 11 years ago

            Quote fom my post:

            q[

            • derFunkenstein
            • 11 years ago

            stop being stupid – I know it’s hard. He did admit flash memory will hit a wall, but it’s farther down the road. Should we stick with old technology just because ti’s the best it’s ever going to be? Hell no.

            • MadManOriginal
            • 11 years ago

            wtg Krogoth, you’ve moved up from mere memes and pictures!

          • Krogoth
          • 11 years ago

          Exponential growth always “comes” to an end due to physical limitations. Flash is not safe from it either. 😉

          I wonder what fabrication process they are using for latest chips? If it is not far from what CPUs and GPUs use then flash may hit a limit sooner then we expect.

          Magnetic media has not come to a dead end yet in terms of data density. It may also experience a revival in performance growth if somebody develops a solution that moves away from the mechanical element.

          5 years is way too optimistic, especially when the economy for most of the world is slowing down.I give it at least 10-15 years for widespread adoption to happen.

          It is more likely within five years that flash drives may start to show up as an option for mainstream desktops (note that it is not enthusiast rigs we are talking about here). Majority of laptops sold will most certainly have them as a default choice.

            • UberGerbil
            • 11 years ago

            Samsung and Toshiba are sampling flash at 32nm; their current production is at 43nm. The Micron/Intel JV apparently has been in production at 34nm for a couple of months now.

            And yeah, they’re hitting the same walls as everything else, and having to go to the same tricks (double-patterning, etc) to get over them. Flash is easier to fab than logic, so it’s not quite the same problem as they have with CPUs, but physics is physics and it is a harsh mistress. There’s a lot of research going on to find the tech that comes next, because everybody thinks the last shrink will be coming soon. PCM (phase change memory) seems to be the favorite, but there are a bunch of techs out there for non-volatile memory — nanowires, etc.

            • MadManOriginal
            • 11 years ago

            Yeah there’s the HP memristor too. I didn’t mean to just focus on shrinks or flash chips, but in any case I have a feeling that the flash chips themselves make up a small part of SSD cost, at least for MLC drives. Also for desktops they could always make drives in 3.5″ form factor as a simple way to increase capacity.

            • Imperor
            • 11 years ago

            Exactly! A 3,5″ Enclosure has a greater volume than 4 2,5″ drives so already we could build 2TB SSD drives. Unless they’ve come up with bigger than 512GB 2,5″ SSDs? There are already at least 1,6TB 3,5″ SCSI 320 drives as well so SSD is definately up to par with mechanical when it comes to capacity!
            What’s left is reliability and price. That shouldn’t be a problem given some time!

            I’m sure mechanical drives will stick around and improve for quite some time yet though, we’ve got SATA 6,0Gb on the way which will do some good and I don’t think the very limit of magnetic storage tech is reached yet by far. Just look at the step up from DVD to BD for instance, what about a similar step for HDD platters? Might not be doable in the same way but there’s a lot of relly smart people out there just thinking stuff up!

    • DrDillyBar
    • 11 years ago

    Pretty cool, too bad it wasn’t released a year ago.
    Still cheaper in the end to just buy lots of RAM and slice off a RAMDrive in Software.

      • indeego
      • 11 years ago

      And get a really really good battery/UPS systemg{<.<}g

        • muyuubyou
        • 11 years ago

        That never hurts.

      • WareWolf801
      • 11 years ago

      Software based ram disks suck from my experience..

    • d0g_p00p
    • 11 years ago

    As usual it’s the price that will keep me away. I mean $400 for this thing, not including RAM? if it was priced at $99 or even $199 I might think about it. I could get 4 OCZ SSD drives for less price and have 200+ GB of flash storage, plus crazy speed if I run them in a stripe.

    Nifty toy, but too expensive and pretty useless (for me)

    • Krogoth
    • 11 years ago

    This drive is a bad and old idea. I like it better when it was called “RAMdisk”. It is made in a time where HDDs for practical purposes and intend have reached a performance plateau.

    DDR2 DRAM should be several magnitudes faster. I suspect the problem to be a combination of OS not knowing how to properly utilize it, poor implementation, and the interface itself is bottleneck.

    Using a volatile storage media for “permanent” data storage is just asking for trouble, no matter what power backup solution you have. RAID 0 version is like playing around with Nitroglycerin.

      • Prodeous
      • 11 years ago

      Yes, but the performance is explosive 😉

      • MadManOriginal
      • 11 years ago

      One problem with this type of device coming from a vendor which doesn’t make memory is that it’s hostage to available DRAM modules and pricing. What if Micron or Samsung made one that had RAM built in? I guess they’re banking on SSDs though because their flash production is worthless otherwise. In addition installation size needs keep increasing for OS and programs with flash SSDs squeezing its usefulness too, 16GB isn’t quite enough to make it compelling. This type of thing was a much better idea before flash SSDs.

      The flash as backup is a bad idea too. I understand they did it to keep it ‘all-in-one’ but I can get a 7200RPM 2.5″ HD for <$80 that would have excess capacity and backup way faster than CF.

        • ludi
        • 11 years ago

        AFAIK the traditional driver of Flash has been handset devices and other portable electroncs, not SSDs. Even without the emerging SSD market, Flash is far from “worthless”.

        The reason Micron et al are very gung ho about SSDs is that they have the opportunity to take over a large, existing market using a product they already make and understand.

          • MadManOriginal
          • 11 years ago

          The second point you make is certainly valid. I wasn’t thinking as long-term as I suspect you were with your first point though, but rather looking at the market over the last year or two for bare flash which has had the rug pulled out from under it even before the world economy hit the skids. I remember lots of talk about flash being non-profitable so being able to bundle it in to an SSD at a large profit is attractive to the flash chip makers.

          I certainly didn’t mean that flash was traditionally driven by SSDs, I’m not sure how you got that.

      • KinCT
      • 11 years ago

      RAID 0 is fine as long as you know and accept the risks. Ages ago, I built one on a Promise IDE controller “modded” to be a RAID controller (added a resistor and flashed the firmware). Strapped together two WD 13 GB drives into RAID 0. Worked like a charm – I had it for seven plus years and then gave it to a friend who is still using it. But yes, to an extent you are playing with fire. But hey, we’re geeks and that’s what we do, right?

        • Kurlon
        • 11 years ago

        Except in this case, RAID 0 is no more dangerous than non RAID, as you’re not adding drives / more moving parts and electronics / points of failure to the equation, you’re just increasing the one drive’s available bandwidth to the computer.

          • Krogoth
          • 11 years ago

          No, you still have to worry about DIMM failure on the other “drive”. Not to mention when the DIMMs go bad, the data in question is completely borked.

            • Kurlon
            • 11 years ago

            A DIMM failure is a DIMM failure, no matter if you’re running single or dual SATA/RAID 0 on this unit. The odds of failure don’t change if you go from single connection ‘SLED’ to dual connection ‘RAID 0’ mode, and neither do your chances of recovery.

            If you’re looking to use this as long term storage, you’ve missed the point of the drive. This thing’s primary use is as temp space, where it can be blisteringly fast. I experimented with the Gigabyte I-RAM back when it was the cool toy on the block. Using it to hold qmail’s queue directory, I could replace 6 machines using 6 disk RAID 5 setups for OS and qmail with one box using a mirror set for the OS and all of qmail’s data save for the temp queue. The I-RAM was lightyears ahead of the RAID arrays for perf in the queue dir, which consisted of lots of random reads and writes, I never could get a traditional RAID going that would let me be CPU bound, with the I-RAM I could. At risk, undelivered mail at that moment. If it did mess up, I’d loose about 3 minutes of mail, which is essentially nothing, so worth the risk to me.

            Where these will really take off is with filesystems that can use the RAM drive for transient scratch space, while streaming a journal to traditional media. You get the speed of the RAM drive, with the reliability of a traditional RAID of magnetic media that’s being used optimally by virtue of the journal being a constant contiguous write op.

            • Krogoth
            • 11 years ago

            In other words, it is a niche product marketed towards the wrong crowd? FYI, I only said that drive was worthless if it is used as “permanent” data storage. It is handy for rigs where upgrading the main memory capacity would be cost-prohibitive or impossible due to slot and MCH-limitations.

    • ludi
    • 11 years ago

    For the money I would take the X25-E. It’s more than fast enough, smaller, and cooler.

    Where this could be useful, if money were no object, is for a dedicated swap-file drive. 8GB RAM would be plenty and cheap enough, and there are no wear-out issues to deal with.

      • MadManOriginal
      • 11 years ago

      The X-25E also has the same capacity – 32GB – from the get-go, and that’s assuming you can even find 4GB DDR2. If you can that RAM alone would run $800+. And a 64GB X-25E is coming and will likely cost a bit less than this thing. would with 32GB.

      They need to make a DDR3 version of this with backup to a HD rather than to add-in flash in preparation for cheaper DDR3 that’s coming. One advantage of that is DDR3 is designed for denser modules, another is lower power draw. If it stays powered on during sleep and does backup to the HD during operation or as part of the sleep routine that would be nice.

        • Vaughn
        • 11 years ago

        I really like that idea ManMan, I think backing up to an HD would be a great addition to a second revision of this product.

    • shank15217
    • 11 years ago

    SATA 3 standard needs to come out desperately.

      • continuum
      • 11 years ago

      Indeed. And it’s coming out exactly for drives like this… well, SSDs mostly, but drives like this too– because they can benefit from the additional bandwidth NOW as opposed to some time in the future. =)

      Pity 4GB DIMMs are still so expensive…

    • MrJP
    • 11 years ago

    This was looking very good until we got to the bit with the price. Seems very expensive for one controller chip, eight RAM slots, a CF interface, SATA & power connectors, a battery, a PCB, and a box. I suppose that’s a reflection of the development costs, but if this ever became popular and achieved real mass production, then I don’t see any reason why it should cost more than a mid-range motherboard. At that kind of price it would become very attractive, but as it stands it would be an odd choice to take this over a flash SSD for less money.

    It would also be an improvement if it could use standby power to allow the disk to remember its contents with the system shut down, with then just a smaller battery supplied for emergencies (does the write-to-CF function work on battery power?).

      • TurtlePerson2
      • 11 years ago

      I thought this was really interesting until I read the price as well. I guess they figure they’ll only sell it to corporations, so why not jack up the price?

      • Kurlon
      • 11 years ago

      There is an external power supply available for these to supply standby power according to HyperOS. And if you think this is expensive, you should have seen the prior generation HyperOS was selling, $2k for the base unit sans ram…

      • designerfx
      • 11 years ago

      It’s good to see this though, expensive development eventually trickles down the hill, and then we all benefit.

      If ram and the devices end up less price and faster than a SSD, you know where the business will be. This is a good competition and should hopefully drive intel and acard to innovate further.

        • clone
        • 11 years ago

        I just don’t like the idea of having power draining all of the time…. any interruption and it’s gone but once I started reading the article I was bouncing in the background how cheap DDR2 667 is and comparing the number of dimm slots…. then I was calculating how much space XP really takes becuase this option would be unsuited for Vista which requires far too much space and would require a notable amount of time spent disabling many of Vista’s newly worthless features if it was used…… precaching, prefetch and so on.

        anyway at this point I hadn’t read about the price of the assembly itself yet…. I had come up with a number for it’s value to me personally which was no more than $100.00 becuase the ram would have to be considered on top of the reliable power requirement as well but then the big reveal and I had to laugh….. $380.00= deal killer…. even at $250 it’s a horrid option.

        yes it’s fast but it’s nearest competition doesn’t need constant power….. this alone should knock 50% of it’s value off…… I’m not sure where this device should be situated in the market given it’s horrible pricing and constant nannying.

        for corporate why not go Intel….. for enthusiasts where this part belongs it’s just not viable… it’s more like a step backwards given the steady state power requirements and absobent cost……

        I like all of the now attainable options and can only hope Intel’s drops by 50% in price at which point I’ll buy one.

          • ludi
          • 11 years ago

          Sir, put down the ellipses and back away slowly.

            • Meadows
            • 11 years ago

            Oh, you beat me to it.

          • Kurlon
          • 11 years ago

          I suspect it’s only a matter of time before they tweak the firmware to trigger an auto backup after X minutes of no power with a restore on powerup. It can already boot while restoring so that half is dealt with.

          That said, if your machine spends much time off… you’re not the target audience.

          • Meadows
          • 11 years ago

          “require a notable amount of time spent disabling many of Vista’s newly worthless features if it was used… prefetch and so on”

          This is one of those comments that bring me close to exclaiming “your mother”. The feature you mentioned is as far from being worthless as possible. Then again, this was expected from a person who usually types 4-8 paragraphs conveying his lack of understanding.

            • Usacomp2k3
            • 11 years ago

            “newly” was the key word in there. He’s saying that since it doesn’t have a traditional, mechanical hard drive, they aren’t as useful any more on a flash drive.
            I did a double-take as well.

            • clone
            • 11 years ago

            meadows I don’t ask for your input, I don’t center out your posts, and I don’t pretend to be special at your expense.

            your an asshole by any measure and you can gfyrslf.

            • Meadows
            • 11 years ago

            I warned you above, and you asked for it.
            Your mother!

            🙂

            • clone
            • 11 years ago

            Meadows being an asshole doesn’t make you special and lacking comprehension skills only makes you an idiot…….the next time you see the word “newly” in front of “worthless features” you could stop and use the pea floating in the case of fluid on your shoulders to understand what’s said instead of proving that you’re an infant fool crying & sobbing profusely to your betters that you aren’t.

      • Bion1c
      • 11 years ago

      same thoughts same thoughts..

      funnily enough $100 was my price limit as well. higher than that and it is not an attractive option.

      still, it is an extremely interesting product. the storage market is really getting fun with SSD tech and now this. I’ve never really understood why the DRAM-as-SATA-disk never took off – yes the non-volatile thing is an issue but the performance promise is tantalizing

      This device is a lot more practical than the i-ram, but its not there yet. It needs a power supply so you can run it indefinitely (this may be available but wasnt mentioned in the article). The backup to CF card is a good concept though. If i was using this in a system I would still take an image of the base system though to restore it when it inevitably lost its data.

      so yeah, basically awesome concept let down by a crappy storage controller and price 🙁 the margins on this device must be huge, and yes they are obviously recouping R&D costs, but at this price point this will be a guaranteed niche product.

      take a risk guys, release it at $100 and look at the sales numbers……

      oh well. looks like we’ll get fast SSDs before anyone can release a compelling ramdisk product 🙁

        • clone
        • 11 years ago

        it’s certainly interesting and a UPS could serve if it was powered independently but the whole idea puts me off…… it’s not that it’s not do-able it’s just not worth it above $100.00

        I agree though this products shelf life will be limited to advances made with SSD’s as their inherrent benefits make this speedy option more antiquated over time.

        hopefully Windows 7 will come with a few predetermined settings that can detect which storage controller is attached and disable features that are unnecessary when a flash SSD is installed.

        I was surfing yesterday and saw some Flash storage 64gb models (G.Skill) for $170 and was considering getting 2 RAID’ing them but am worried about how reliable they would be given most are unstable with multiple IO’s except Intel’s.

    • Dposcorp
    • 11 years ago

    Excellent article Geoff.

    I have some questions and a comment.
    Sorry in advance for the long post.

    Question is, did you guys try the back up and restore feature?
    If so, how did it work? Does it do a sector by sector copy?
    Would the Compact flash drive be bootable in a emergency?
    Is the backup done via some Drive Imaging software that comes with it, or some app stored in the firmware?

    I was thinking that a 16GB (8x2GB) Boot/OS would be nice, and a 16GB CF or SDHC card inside of a CF adapter would make a a good back up. I could all my important docs in something the size of CF card.

    I’d like to comment on the warranty issue with solid state drives. Correct me if I am wrong, but most SSD have a 1 to 3 year warranty.

    Most DDR2, CF & SD cards have a lifetime warranty. Now I am not saying I would use these things for the next 20 years, but 4-8 years is something I can easily for see. I know lots of people running systems as their main or secondary machines that are 4-8 years old.

    The ANS-9010 gives you the option of a longer warranty, and as well replacing only parts of the drive that need it, like 1 or 2 sticks of ram at a time, and you could even use mem test to easily verify the problem sticks.

    Same thing with your back up flash cards…..a simple email and rma and you get a new one. Not to mention that you’ll also be able to use them in other devices. I have seen 16GB SDHC cards for at newegg for $24, and 16GB micro SD cards at buy.com for $16 (although Sandisc only gave them a 5 yr warrant).

    Having a micro SD card that can be booted in a emergency with all your most important stuff is pretty cool. Add in the fact that when you buy a larger card, the smaller ones can be used in phones, cameras, music players, etc, and I can see a lot of versatility in a system like this.

    This price is a bit high now, but I can easily see it coming down to a $500 or $400 for the unit, 16GB of DDR2, and a 16GB flash drive. Not a bad deal for a fast drive with no moving units, its own method of backup, and easily replaceable components with long warranties and multiple uses.

      • Kurlon
      • 11 years ago

      The prior generation setup HyperOS was selling would do automatic backup to 2.5″ HD or CF, and would allow boot/initial operation off the backup while it restored to ram live. I’m guessing they cut that functionality as a way to lower the cost.

      Based on HyperOS’s available info, the backup is done via the firmware, not software.

      What firmware rev did TR use for the review?

      • Kurlon
      • 11 years ago

      Looking at the beta firmware release notes, ECC RAM support is now listed. Read perf has been improved in RAID, and you can now boot while restoring in RAID mode, implying you can do live restore and run at the same time.

    • pedrofdmp
    • 11 years ago

    Serial ATA is not 300 MB/s but 3.0Gb/s

    3000 Mb/s megabits/second is 375MB/s megabytes/second (3000/8=375)

    note lower case b goes for bits as gigabits Gb, upper case B goes for bytes like megabytes MB

    Serial buses like serial ata are always measured in bits 1,5Gb/s 3,0Gb/s

    Paralel interfaces like ATA are measured in bytes 133MB/s 100MB/s …

    Please correct the article.

    §[< http://en.wikipedia.org/wiki/Sata#SATA_1.5_Gbit.2Fs_and_SATA_3_Gbit.2Fs< ]§ §[<http://en.wikipedia.org/wiki/DDR2_SDRAM#Specification_standards<]§

      • emorgoch
      • 11 years ago

      Read up on the standard.

      SATA has a transfer rate of 3.0Gb/s. However, it also uses a 10b/8b encoding (http://en.wikipedia.org/wiki/10/8_encoding). The end result is that for every 1B of disk data that it read/written, 10b flow across the wire. The end result is that only 80% of the data on the line is actual user data.

      Thus: 3.0Gb/s * 0.80 == 375MB/s * 0.80 == 300MB/s

        • equivicus
        • 11 years ago

        l[

      • eitje
      • 11 years ago

      oh, you owe someone an apology!

    • bittermann
    • 11 years ago

    Ok, stupid question but I’m going to ask it anyway. I see from the picture that they were using 5-5-5 DDR2 ram. Would having faster 4-4-4 DDR2 ram help in any way or would the bottleneck still be I/O of the bus?

      • eitje
      • 11 years ago

      Looking through the list of supported memory, it looks like 5-5-5 is the expected timings for the RAM to be used.

    • Prospero424
    • 11 years ago

    Just out of curiosity: why were these tests run on a Pentium 4 platform with an ICH7R?

    Why not the modern hardware used in other articles/reviews? Seems like the results would be a bit affected.

    Is it to keep the hardware constant from the older SSD reviews/results?

    • lhl
    • 11 years ago

    Hmm, 20K IOPs isn’t bad, but if we’re talking about $1200 for 32GB… It’s actually /[

    • Prospero424
    • 11 years ago

    /[

    • inphobia
    • 11 years ago

    hey, an el-cheapo version of the ramsan 🙂

    §[< http://www.superssd.com/<]§

    • hermanshermit
    • 11 years ago

    The best solution as RAM modules become larger is to allow BIOS partitioning of RAM for system and HDD use. No additional hardware, no restrictive SATA needed and could support powered retention and booting.

    In 2-3 years is Windows 7 really going to be able to use 6*8GB of RAM in 3 channel mode?

      • Anonymous Coward
      • 11 years ago

      An OS could split the RAM up easily enough. Right now the trouble is that masses of RAM isn’t cheap enough, largely due to there not being enough slots on the motherboards.

      There are, however plenty of PCIe slots available. I think it makes excellent sense to somehow get extra RAM hanging off of those. This could be “L2 RAM” which the OS uses for file caching and swap space.

        • bittermann
        • 11 years ago

        “An OS could split the RAM up easily enough. Right now the trouble is that masses of RAM isn’t cheap enough, largely due to there not being enough slots on the motherboards.”

        Now days most come with at least 4 ram slots which could equate to 8/16Gb’s of ram. That would be plenty for swap file, caching, etc…plus with newer Phenoms you have dual memory controllers. 4gb Unganged memory for the OS, 2-3gb fast swap file and 1-2gb left over for cache…now that sounds sweet!

    • Prodeous
    • 11 years ago

    I was really hoping that this device would be at least twice as good as Gigabyte I-Ram. I guess I should not expect miracles.

    Outside of 3 improvements, its just an I-Ram.

    1. higher capacity
    2. CF backup

    dual sata – raid possibility, an interesting toy, but it really depends on the performance of on-board raid controller. As someone mentioned, it would be nice to see a test with a professional raid card, maybe the on-board raid is not simply efficient enough to bring out the true performance of the device.

    with regards to comments about PCI-e device. that would be a mistake. what this device, and Gigabyte I-Ram drive bring is totally pass through device, what i mean, is no DRIVERS. a pure SATA drive from the systems perspective. having a PCI-e device would require drivers, etc etc… there are plenty of devices like that. true expensive, and a cheaper solution would probably be welcomed.

    Just to correct those of you who are thinking that the Gigabyte i-ram PCI device used PCI to transfer. The Gigabyte I-Ram used it only to take power, and standby power when the system is off (but pluged into the wall). As there is enough power going down the PCI lanes to keep the ram active. I would like to know how this device keeps its ram powered when the system is shutdown? if it is just from standard SATA power cable.. does SATA cable keep energy flowing when system is shut down? Or is it just with the battery pack?

    One thing they could have implemented is on-board RAID0. Granted that would saturate the SATA cable, but then it would not have to rely on south bridge to perform such functions (as it could be holding down the performance).

    There is plenty of possibility in this device.
    Maybe in Version II (if there is one) they will sort it all out, but i guess till then, my Gigabyte I-ram will just have to do.
    Maybe Gigabyte will release their II-RAM, or whatever ther DDR2 prototype was called.

      • tjoynt
      • 11 years ago

      I think they’re using raid-0 so they can use two SATA connections: they are already saturating one. If the southbridge doing raid-0 was the bottleneck, you would get better performance from a single SATA connection, which is not the case. AFAIK, there is no reason internal to the ram drive to use raid-0, just the need for more bandwidth to the rest of the computer.

    • Fighterpilot
    • 11 years ago

    Very nice review…thanks TR!
    You have to bow to anything that wins basically every test handsomely against the high performing,popular parts we currently use.
    With almost nonexistent seek times and a fresh Windows 7 install it would be cool to see how fast the desktop would perform.

    • mako
    • 11 years ago

    Interesting product. Any chance of popping the heatsink and seeing what sort of FPGA they used? If not, can you tell if the chip package is metal or organic?

    • MadManOriginal
    • 11 years ago

    It is neat but it’s too little too late. Flash SSDs are coming on too quickly and even the most expensive ones have this beat in $/GB plus this drives max capacity with semi-sanely priced DDR2 isn’t great. DDR3 is also going to become mainstream quickly and drop in price so dDR2 will go the way of DDR1 and possibly rise in price. If it had come out a few years ago it might have been nice, although DDR2 was way expensive then, but now…meh.

    • Meadows
    • 11 years ago

    Now here’s what I call a bad storage controller.

    • Anonymous Coward
    • 11 years ago

    Someone should definately make a PCIe RAM disk and be done with it. No SATA interface crap. Make the battery optional, and kill the flash backup. Solder the RAM onto the board. Simplify and cheapen.

    Its not too hard to make a script to back up files to another internal hard disk, if needed. It would take the Linux server crowd just a moment to whip up code to use such a device as a magic file caching space (no backup needed). It would be awesome to go buy 64G “file cache cards” to plug in without any configuration needed. Imagine whole databases and web applications living in there. Would be great for read-limited applications.

      • Saber Cherry
      • 11 years ago

      *[

      • d0g_p00p
      • 11 years ago

      It’s been done already. Fusion-IO

        • UberGerbil
        • 11 years ago

        And it requires special drivers which for a long time didn’t support using it as a boot drive. There’s something to be said for standards and looking like commodity hardware to the OS, even when you’re not.

          • d0g_p00p
          • 11 years ago

          Sorry but you are wrong. I have 2 of them. Both are in production. it’s true that you need drivers to boot off the thing. However I don’t use it for that. I have a final and a prototype running my production and test web servers. neither required drivers and can be seen in the BIOS as a standard drive.

        • Anonymous Coward
        • 11 years ago

        They make only flash storage as far as I can tell. Also, as far as I can tell, its expensive, fairly small market, and proprietary. Useless!

          • ludi
          • 11 years ago

          Uh, but doesn’t “expensive, small market, and proprietary” describe most of the device you just asked for?

          Soldered down memory isn’t necessarily cheaper because not everyone is going to want the same amount, which means predicting and stocking multiple inventory items and possibly guessing wrong — since the market is small and there isn’t a lot of data to guide the feature-slotting decisions.

          Script-based backup is a proprietary feature that not everyone can conveniently use, while a CF card and a button are pretty easy. And a SATA connector and controller ensures that any modern system can recognize the device as a disk storage entity, no OS-specific support required.

    • Vasilyfav
    • 11 years ago

    Once again proves that RAMdisks are only useful for their IO throughput and nothing else.
    Too expensive, volatile, not enough performance difference otherwise.

    • sergal
    • 11 years ago

    I’d like inform you that Scarlett Johansson “actress”actually is a clone from original person,who has nothing with acting career.Clone was created illegally using stolen biomaterial.Original Scarlett Galabekian last name is nice, CHRISTIAN young lady.I’ll tell more,those clones(it’s not only 1)made in GERMANY-world leader manufacturer of humans clones,it’s in Ludwigshafen am Rhein,Rhineland-Palatinate,Mr.Helmut Kohl home town.You can’t even imaging the scale of the cloning activity.But warning,H.Kohl staff strictly controlling their clones spreading around the world,they’re NAZI type disciplined and mind controlled,be careful get close with clones you will be controlled too.Original family didn’t authorize any activity with stolen biomaterials,no matter what form it was created in,it’s all need to be back to original family control in Cedars-Sinai MedicalCenter in LA.Controlling clones is US military operation.Original Scarlett never was engaged,by the way

      • Farting Bob
      • 11 years ago

      I want some of what you’ve been smoking!

        • paulWTAMU
        • 11 years ago

        I can’t tell if he’s a troll or mentally ill.

      • Krogoth
      • 11 years ago

      Go away damm spammer bot.

        • derFunkenstein
        • 11 years ago

        with no link, I can’t believe he’s spamming. Unfortunately, you’re dumb, so I wouldn’t expect you to understand.

          • lethal
          • 11 years ago

          nah he is right on this. Just google a quote from the insane post. Its been copy pasted on multiple boards/forums verbatim.

      • eitje
      • 11 years ago

      I think there’s a hidden message here…

      • equivicus
      • 11 years ago

      Well now that Nazi has been mentioned this thread is dead. 🙂

    • emorgoch
    • 11 years ago

    Geoff,

    Decent review. I definitely believe that there’s some performance to be had out there if they used a better controller. However, I find your usage of the RAID-0 mode puzzling. While motherboard RAID controllers get the job done, they aren’t exactly known for their efficiency. While they may be good enough for mechanical drives, I can’t help but wonder if they’re turning out to be a bottleneck in the case of an SSD RAID, or the ACard.

    I would be truly interested in seeing what the performance would be like using proper RAID card in the system. I know it would invalidate all the previous hard drive performance numbers, but even just testing it out to see if the on board RAID was causing a performance hit.

      • Anonymous Coward
      • 11 years ago

      A proper raid card for raid 0? I have to agree that there is a risk of the crappy onboard raid controllers being too slow, but I am horrified to think of people having to buy expensive raid controllers to run raid0. This is a perfect application of pure software raid.

    • glynor
    • 11 years ago

    It’s obvious that the controller in the drive is holding this back, like those early SSDs.

    It’s really too bad the device itself is so expensive. I wonder how much of that is designing for using DIMMs vs greed vs the actual cost of the components.

    If this was built differently, I think it could be quite good! Imagine if Intel built a version of their drive, with DDR2 chips instead of Flash. Not a novelty “use DIMMs in it device” but a 3.5 in 32GB (or 64GB) drive built with their controller out of DDR2 chips! In fact, combine it with the “old” idea of hybrid drives and include a tiny magnetic 64 GB platter in the drive and have it run in kind-of a delayed write “mirroring” kind of mode with the DDR2 chips (for power outages) and a small user-replacably lithium (or LI rechargable, even) battery for short spells.

    That could easily change the meaning of high performance for storage.

    • odizzido
    • 11 years ago

    Unless these move to PCIE I can’t see ram drives having much of a future with SSDs on the way.

      • Saber Cherry
      • 11 years ago

      I said that, just with more words. I could expound upon how ‘brevity is the soul of wit’ for hours…

    • indeego
    • 11 years ago

    /[<"and there's less reason to worry about the drive eventually wearing out."<]/ Alas, one should look at bit flips/soft/hard errors that exist to a much greater extent with this type of setupg{<...<}g

      • Saber Cherry
      • 11 years ago

      Why should ECC on DRAM be worse than ECC on flash? Bear in mind that data has to sit around on flash for much longer, while it is constantly being refreshed on DRAM, so errors can’t accumulate in a single bit until it flips; it has to be flipped by a single incident. And DRAM doesn’t have the fundamental problem of being burnt-out by use, so a similar level of ECC would be more protective to DRAM. Also note that I don’t know where the ECC in DRAM actually occurs, so this is an actual question, not a sarcastic response. Non-parity RAM uses ECC somewhere, but maybe only for signaling and not for storage… I just kind of assume that parity RAM is only an extra level of precaution for the errors that are not detected by the first and second lines of ECC defense inherent to DRAM, but maybe those lines of defense are only in my imagination.

      I’ve never really heard of actual problems from RAM errors and consider them more theoretical, except in space. Though I suspect supercomputer clusters and server farms might notice them.

        • data8504
        • 11 years ago

        ECC is provided by extra DRAM capacity on the actual DIMM. A scrub is performed at every refresh, and transparent errors are only optionally recorded in the system NVRAM error log. There are many schemes, including simple parity, CRC, etc., but you are completely right that the risk of “non-transparent” correctable errors is much lower on DRAM, theoretically because of the more atomic requirement.

        Though, I might add – as far as I am aware, a lot of flash has no ECC capacity embedded into it. HDDs don’t either, but in an odd twist, CDs (and DVDs, I believe) do.

        As an aside, the 10b-14b encoding on CDs (allows derivation of clock, control symbols, etc.) is crucial in cheap CD-Rs because bits begin to fail very quickly, and the encoding becomes a quite necessary crutch.

          • Saber Cherry
          • 11 years ago

          When you say ‘extra capacity on the actual DIMM’, do you mean as an extra chip, or in an itty bitty chip elsewhere on the PCB, or what? Because I just counted one of my old DDR1 sticks and it has only 8 chips (but 9 spaces for chips, I suppose for the parity version), so should I conclude that it has zero ECC for actual storage, or just that it has ‘very little’ (like enough for a single parity bit per row)?

          DVDs have less space dedicated to ECC than CDs (as a proportion of capacity) but it has greater capabilities (using PAR2, IIRC, though I don’t know how that works anyway). I vaguely remember a 1/3rd vs 1/7th of capacity devoted to ECC, otherwise scratches and dust would be terminal problems. 1/3rd seems to agree roughly with the 10/14 bit scheme you mentioned assuming there are a few more bits used for larger blocks.

          Data CDs have more ECC than audio CDs (and you can in fact burn more data on them if you force the data to be burnt as audio, though it’s unreliable and unreadable by normal software). I’m not really sure about audio/video DVDs versus data DVDs.

          As for HDDs lacking ECC – I was not aware of that, but ECC is still present (in software) in the file system. Though I suppose technically a person could use a file system that did not have ECC. Actually, I guess file systems would inherently implement some degree of software ECC on a flash drive anyway, even though a competent company should include it in hardware.

            • Bauxite
            • 11 years ago

            You answered your own question, the real ECC dimms have 9 memory chips. Most people don’t have these…honestly less than 1% is probably accurate.

    • Saber Cherry
    • 11 years ago

    Let me begin by stating that I would absolutely buy one of these (and RAM) if I had $500 laying around looking for a home. That said –

    I’m glad to see this area is still being actively developed. I find slow HDDs to the most annoying aspect of modern system performance (excluding unpredictable externalities like the internet), and currently, it’s just hard to take advantage of cheap memory, with predominantly 32-bit programs and (personally) a 32-bit OS anyway. What I really want is just a place to put my swap file to get huge, fast virtual memory (in the literal sense).

    Unfortunately… while their implementation /[

    • Hance
    • 11 years ago

    Cool toy but it would cost a fortune to make it usable on top of the already hefty price for the drive itself. I can’t even find a 8 gig dimm on newegg the biggest they have is 4 and they are about 100 bucks. A 32 gig CF card for backups is another 75 bucks. By time you add up the drive, memory and a cf card for backups your out 1200 bucks.

    Its a nifty toy but its not 1200 bucks nifty.

      • BobbinThreadbare
      • 11 years ago

      4 gigs would be enough for your page file, 8-12 would be enough for windows, and a couple common applications. You couldn’t put everything on it, but I wouldn’t say it has no use.

Pin It on Pinterest

Share This