SilverStone HDDBoost Review

[AkG]

​

SilverStone HDDBoost Hybrid SSD / HDD Device Review

Manufacturer's Product Page: SilverStone Technology HDDBoost
Price: MSRP $49



There really is no denying that solid state drives have a level of power and performance in certain crucial areas which simply make them better than hard disk drives as “Operating System” drives. With the number of SSD reviews showing their faces around the internet, it should be obvious to you that even the slowest modern generation solid state drive is faster in most areas than the fastest hard drive. However, all that power and performance comes with a massive increase in price and a huge decrease in storage capacity versus a spindle-based drive. This has led to many people forgoing the SSD upgrade option for the time being.

In the past, hard drive manufactures like Seagate and Samsung have tried to combine the power of solid state drives with the capacity of spindle-based drives with their “hybrid” products. Sadly these efforts were less than impressive and were quickly discarded. Meanwhile, Adaptec has approached this problem differently with their MaxIQ kit MaxIQ kit , which allows you to use a solid stat drive as cache for a raid array. Unfortunately, this solution is designed for an enterprise environment, costs a heck of a lot of money and is limited in which SSDs you can use.

Users want the best of both worlds: the storage capacity of a spindle-based drive with the speed of an SSD and this is where SilverStone’s HDDBoost comes into play. The HDDBoost allows you to mix and match different hard and solid state drives while combining the units in such a way that if one drive kicks the bucket, the data on the other is still safe and sound. In a nutshell, the HDDBoost is a two port RAID 1 controller with a twist and besides the obvious limitation of one HDD and one SSD, allows for a massive amount of flexibility. We will get into what makes the HDDBoost so unique, but needless to say this is certainly a novel and intriguing way of overcoming past (seemingly insurmountable) problems which plagued earlier Hybrids. The ability to chose your level of performance and storage size based on your own personal criteria is seductive, but only if it works. SilverStone also claims that by minimizing the writes to the SSD, you will effectively extend its life as well.

This unit doesn’t include a drive but should rather be treated as the means to combine two dissimilar drives one virtual drive. As such, it comes with a frugal price of around $49. With a price such as this, the HDDBoost is intended to bridge the chasm between consumers that can afford high performance, high capacity SSDs and those that just want some extra speed in their system without sacrificing storage capacity.

To help us figure out how this product performs in its intended market niche, we have carefully selected two moderately priced drives to showcase the HDDBoost’s abilities. On the hard drive end of things we have chosen the venerable Western Digital 320GB single platter hard drive which goes for about $45, and for the solid state drive we have gone with the Kingston branded version of the Intel X25-V 40GB: the now defunct Kingston SSDNow V 40GB which retailed for as little as $90. All in all, this could prove to be an interesting review.

​

 

[AkG]

Specifications & Features

Specifications & Features

​

 

[AkG]

SilverStone's HDDBoost Technology Explained

HDDBoost explained

<img src="http://images.hardwarecanucks.com/image/akg/Storage/HDDBoost/specs_2.jpg" border="0" alt="" />
​

When we first heard about the SilverStone HDDBoost we were tempted to write it off as nothing more than a modified RAID 1 controller with a solid state drive added into the equation to boost its appeal. In some ways this is exactly what SilverStone has done, yet in many ways misses the true nature of exactly what is going on behind the scenes. In fact, it is closer many regards to say that of Microsoft’s ReadyBoost with some RAID 1 abilities added in for flavouring, rather than to call it a RAID 1 controller. Before we get to testing whether or not the theory and execution of this interesting little piece of technology actually works in the real world, lets first take a look at exactly what the HDDBoost is (and is not) and what makes it special.

<img src="http://images.hardwarecanucks.com/image/akg/Storage/HDDBoost/raid-1.jpg" border="0" alt="" />
​

At its heart the HDDBoost is a modified two SATA port RAID 1 controller. Basically, in a standard RAID 1 array the data is read from any drive in the array and written to both drives at the same time. With the HDDBoost, if the data exists on the solid state drive the controller directs the storage retrieval command towards the solid state drive. If the data does not exist on the solid state drive, only then will data be read from the hard drive.

Using this logic means that read speeds can be greatly increased on the most important data: applications and operating system files. However, you are not limited in the size of your “C” drive like you would be with a more typically sized solid state drive. The one downside to this is that data has to be properly placed on the drive as the HDDBoost does not care what pieces are “time sensitive” which are not. Data is data to the HDDBoost.

Setting aside this potential issue we have to wonder how much latency is added into the equation by doing this as you are adding in another layer (and controller) into an already-complicated equation.

<img src="http://images.hardwarecanucks.com/image/akg/Storage/HDDBoost/block_diag.jpg" border="0" alt="" />​

Assuming the HDDBoost works as advertised, this boost in performance is all well and fine for reads but smaller writes happen on a near constant basis in any modern operating system. While hard drives are not nearly as fast as modern solid state drives, older generation drives (such as the JMicron 602 controller based ones) had the colloquial term “stuttering” applied to them. Stuttering is exactly what can and did happen when all the “free” cells on these older SSDs were in a degraded mode and multiple read and writes commands were sent all at the same time. To alleviate our worries regarding the write performance of the HDDBoost and breathe new life into the older generation solid state drives, SilverStone has directed ALL write requests to the attached hard disk. The only time writes happen on the solid state drive is either during a manual update via the HDDBoost software or during reboot. This means that cell usage is extremely light and you would have to reboot your system an average of 10,000 times before cells in the solid state drive would wear out. This also means that write speeds with the HDDBoost will never approach those seen on SSDs but that’s not what this devices is designed for.

<img src="http://images.hardwarecanucks.com/image/akg/Storage/HDDBoost/block_diag2.jpg" border="0" alt="" />
​

Unfortunately, there is a major down side to doing things this way. It may be a lot easier on the SSD’s cells and eliminates stuttering, but it means that unless you either reboot your system frequently or manually trigger (via included software) synchronization, rebooting may take an awfully long while. If the HDDBoost does incremental backups, the added time may not be all that bad while if full backups are done every time you reboot, this could significantly add time to the process. Worse still, the time required would only get worse the larger the solid state drive used.

<img src="http://images.hardwarecanucks.com/image/akg/Storage/HDDBoost/defrag.jpg" border="0" alt="" />
​

Interestingly enough, the controller is not smart enough to just load the OS and most frequently used applications on to the solid state drive. Rather, it clones the first chunk of the hard drive to the solid state drive. This means if you use a 30GB solid state drive the first 30GB worth of data on the hard drive will be copied to the solid state drive, even if this includes less time sensitive data like songs, pictures, etc. Needless to say, you still need to keep you hard drive defragmented or HDDBoost performance will suffer as well.

 

[AkG]

Closer Look at the HDDBoost

Closer Look at the SilverStone HDDBoost

​

As you can see SilverStone certainly did not waste much time or resources on the HDDBoost’s packaging. For all intents and purposes this is a brown cardboard box with details, specifications and a schematic of the HDDBoost printed in black ink on it. Over all it may be bland but it certainly is more than good enough to keep the HDDBoost safe and sound in transit.

The list of accessories is short and sweet and this is par for the course considering the value orientated price niche this unit occupies. In a nut shell you get mounting screws, a thick multi-language instruction pamphlet and even a short, red SATA cable. The only thing conspicuous by its absence was a CD with the HDDBoost application on it. Rather, the applications we will be talking about in the next section are freely available on SilverStone’s website.

​

This is a very basic looking device and upon first glance it looks nothing more than a hot swap 2.5 to 3.5 drive adapter which has been beefed up a bit more than usual.

​

The metal chassis of the HDDBoost has holes drilled into the bottom and sides to allow for securely mounting both the 2.5” SSD inside it and the HDDBoost to your case’s 3.5” hard drive cage. The SSD of course slides into the back of the unit where a standard SATA power and communications ports have been install onto the PCB.

​

Looking at the back of the unit you can see there is not only SATA ports for a 2.5” to 3.5” hot swap SATA adaptor but also a large chip which usually has no place residing on something as simple as a adapter. This is where the simplistic first impression goes off the rails and the secret of the HDDBoost starts to come to light.

​

The controller chip is labelled “SST-HDDBOOST 10005” and tells is us very little about what it actually is, but it does shed a lot of light onto the capabilities and limitations of the HDDBoost. Since there are no external RAM nor NAND chips to be found anywhere on this device, we can make the reasonable assumption that the HDDBoost is going to have very basic cloning functions as there is simply nowhere to store (in a non-volatile format) the information needed for making incremental backups.

When you reboot your system this controller most likely initiates a full clone procedure and if this is the case and it can not prioritize synchronization, reboot times could be down right brutal on even relatively small 30GB solid state drives; no matter how fast your drives are, moving that amount of data does take time. However, there is a good possibility that this System On a Chip does have a trick up its sleeve to prevent this and other issues from popping up. After all it would very easy simply have the SoC give low priority to the synchronization process and even pause it when peak demand happens (which is what happens when you reboot).

If this is the case, the first boot up after installing the HDDBoost should in theory only take as long (or maybe slightly longer due to increased latency from the HDDBoost and its controller) as it did using only your old hard drive. The only down side to doing things this way and prioritizing things in a logical manner is that the longer the system is up and running the less data it will be able to read from the SSD upon the next reboot. This does have the potential to make booting up your system extremely variable with anything from SSD levels of performance to your original HDD level of performance to something in between. We are betting that this is how SilverStone has done things and that most reboots will fall somewhere in between these two extremes.

​

Moving onto the SATA ports we have the typical SATA power port which of course supplies power to both the device and the SSD. The top SATA port connects from the HDDBoost’s PCB to your motherboard while the bottom one goes from the HDDBoost to your chosen hard drive. This in a nutshell ties the SSD, HDD and controller into one big loop with the onboard controller being responsible for what data is sent to the SSD and to the HDD. This is also what allows the unit to clone data from the hard drive to the solid state drive without needing the involvement of the motherboard or your computer’s other components.

​

The other benefit to doing things this way is the HDDBoost and its two devices are seen by your computer’s BIOS as one single drive. Thus there should be no compatibility issues at the physical level with the HDDBoost, or at least none as long as you have room for two drives in your drive cage.

​

At the physical level the HDDBoost is a raid controller, but we can see one minor potential downside to doing things this way. The downside we can see is that unlike true hardware or software RAID controllers (which bypass the motherboard SATA controller and use PCI/PCI-E/etc lanes) or even motherboard software raid this unit shares one SATA rev 2.0 port for both drives. So, as long as you don’t use a high end modern solid state drive which can do 230+MB/s sequential reads, this not going to be a big deal. However, in the future we hope to see this device with SATA 6Gbps compatibility.

 

[AkG]

Software & Synchronization

Software & Synchronization

​

The HDDBoost Utility which you can download from SilverStone’s website is very basic and totalitarian in its approach to things. At its most basic level, this free utility tells you exactly what your HDDBoost is doing at a given time.

Monitoring is definitely an added benefit for this software but unfortunately it has no manual or how to guide to speak of so you are left wondering exactly what some of the options actually do. However, we liked the fact that you can instantly see not only what drives are attached but what their serial numbers are.

​

As mentioned this utility will not only tell you what the HDDBoost is doing but how far along in the synchronizing process it currently is. Its other main purpose of this is to act as a pain free way of updating the firmware on your HDDBoost. Basically, all you need to do is download the firmware file to any attached storage device, make sure the unit is not in the process of synchronization and hit “load” up in the left hand corner. After you go through a few simple steps, it will update the firmware and when finished ask you to reboot the device so the new firmware can be implemented.

​

While there are plenty of good things to say, this is not a perfect application for the simple reason that it is unclear exactly how you can manually trigger synchronization. We assume that hitting the “start” button starts the process and “stop” stops it even though it is fairly obvious what process you are starting and stopping.

The only other minor annoyance we have with this utility is the fact that it does not include a method of polling the SMART data from the connected drives. This is a missed opportunity as you really will have no way of knowing the condition of your drives without yanking them from the HDDBoost and plugging them back into the motherboard.

All in all, we like this nifty little utility but it does need either a FAQ or a simple “how to” guide. It is simple yet effective and above all else it gets the job done.

What is Synchronization?

Upon reboot (or when a user manually triggers a synchronization request via the software), the HDDBoost checks the size of the connected SSD and goes about copying and cloning that same amount of data from the Hard Drive to it. This is much like what your favourite cloning software does, but it is being done at a hardware level much like a RAID 1 array and is doing it seamlessly, without your system’s knowledge or consent. By the same token it is not like a RAID 1 array as the Solid State Drive does not need to be the same size as the Hard Disk Drive.

SilverStone calls this process Synchronization as that is exactly what it is. If you swap in a bigger or smaller SSD, the HDDBoost will simply use the new drive’s size for its clone parameters. It does not care how big the hard drive is as it is only taking (up to) the amount that the solid state drive can store and no more. Obviously, if you have less data on the hard drive than the total amount of storage available on the Solid State drive, it will clone all the data and then stop. This is the theory that makes this such a powerful unit as you get the best benfits and performance of a RAID 1 without the downside of needing a 1:1 storage capacity across both drives.

Of course, once you install a program you are going to have to either reboot or do a manual synchronization to get any of the solid state performance. This is because the programs are still primarily installed onto the HDD and the info needed to allow the SSD to boost performance is only passed over during synchronization. This to us does add an extra layer of hassle factor to the unit, but it is certainly not a deal breaker.

For those of you interested it took about half an hour for synchronization to complete on our Kingston SSDNow V 40GB but the process was completely transparent to the user. Basically, you can use your system as normal while this process is going on behind the scenes. If you reboot the system while sync is happening it will just resume upon your next reboot.

 

[AkG]

Testing Methodology

Testing Methodology

Testing a device such as this is not as simple as putting together a bunch of files, dragging them onto folder on the drive in Windows and using a stopwatch to time how long the transfer takes.

Even when the hardware issues are taken care of the software itself will have a negative or positive impact on the results. As with the hardware end of things, to obtain the absolute best results you do need to tweak your OS setup; however, just like with the hardware solution most people are not going to do this. For this reason our standard OS setup is used. However, except for the XP load test times we have done our best to eliminate this issue by having the drive tested as a secondary drive. With the main drive being a WD 320 single platter drive.

For real world benchmarks we timed how long XP startup took, Adobe CS3 (w/ enormous amounts of custom brushes installed) took, how long a single 4GB rar file took to copy to and then from the drives, then copy to itself. We also used 1GB of small files (from 1kb to 20MB) with a total 2108 files in 49 subfolders.

Please note to reduce variables the same XP OS image was used for all the hard drives.

All tests were run 4 times and average results are represented.

Processor: Q6600 @ 2.4 GHZ
Motherboard: Gigabyte p35 DS4
Memory: 4GB G.Skill PC2-6400
Graphics card: Asus 8800GT TOP
Hard Drive: 1x WD 320
Power Supply: Seasonic S12 600W


SSD FIRMWARE (unless otherwise noted):
Kingston SSDNow V 40GB: 02G9

Drives Used for HDDBoost:

Hard Drive: Western Digital 320GB WD3200AAKS-00B3A0 Single Platter Drive
Solid State Drive: Kingston SSDNOW V 40GB

 

[AkG]

Read Bandwidth / Write Performance

Please note:
Due to the fact most synthetic tests use temporary files only to benchmark performance; all but three have been redacted from our usual review. We believe it would not be fair to include any of these inconclusive tests as they simply test the hard drive portion of the HDDBoost (through the use of temporary files only).

To further help clarify and give as best a understanding of the strengths and weaknesses this device have we have included two sets of data, one before synchronization was allowed to occur and one set for after it occurred.

Read Bandwidth

HDDBoost = WD 320GD HDD + Kingston SSDNow 40GB

For this benchmark, HDTach was used. It shows the potential read speed which you are likely to experience with these hard drives. The long test was run to give a slightly more accurate picture.

We don’t put much stock in Burst speed readings and this goes double for SSD based drives. The main reason we include it is to show what under perfect conditions a given drive is capable of; but the more important number is the Average Speed number. This number will tell you what to expect from a given drive in normal, day to day operations. The higher the average the faster your entire system will seem.

While we are not seeing a 70% increase in performance like SilverStone claims, we are seeing a marked improvement and in fact this device does boost the older, and down right decrepit (by today’s standards) 320GB drive above the magic 100MB/s mark in average read speed, which is nothing to sneeze at.

Write Performance

HDDBoost = WD 320GD HDD + Kingston SSDNow 40GB

For this benchmark HD Tune Pro was used. To run the write benchmark on a drive, you must first remove all partitions from that drive and then and only then will it allow you to run this test. Unlike some other benchmarking utilities the HD Tune Pro writes across the full area of the drive, thus it easily shows any weakness a drive may have.

<img src="http://images.hardwarecanucks.com/image/akg/Storage/HDDBoost/write_proof.jpg" border="0" alt="" />

​

This is very, very interesting; in fact it is so interesting we have included a screen shot of exactly how this came about. As you can see writes, are definitely occurring in real time on the solid state drive portion of this setup. For the time being we are going to mark this down as noting more than an interesting aberration and withhold judgement until the real world testing stage is completed.

 

[AkG]

Random Access Time

Random Access Time

HDDBoost = WD 320GD HDD + Kingston SSDNow 40GB

To obtain the absolute, most accurate Random access time, h2benchw was used for this benchmark. This benchmark tests how quickly different areas of the drive can be accessed. A low number means that the drive space can be accessed quickly while a high number means that more time is taken trying to access different parts of the drive. To run this program, one must use a DOS prompt and tell it what sections of the test to run. While one could use “h2benchw 1 -english -s -tt "harddisk test" -w test” for example and just run the seek tests, we took the more complete approach and ran the full gamut of tests and then extracted the necessary information from the text file. This is the command line argument we used “h2benchw 1 -a -! -tt "harddisk drivetest" -w drivetest”. This tells the program to write all results in english, save them in drivetest.txt file, do write and read tests and do it all on drive 1 (or the second drive found, with 0 being the OS drive).

<img src="http://images.hardwarecanucks.com/image/akg/Storage/HDDBoost/random.jpg" border="0" alt="" />
​

There is no getting around the fact that the single platter version of the Western Digital 320GB drive is a less than optimal drive when it comes to latency. However, even when hobbled with such a less than optimal drive our Hybrid setup does post a much more respectable full drive latency average of 13.41ms. Even more interesting was the fact that while the Kingston SSDNow V 40GB has 0.07ms latency when on its own the “lower 512mb” portion of this test shows the Hybrid setup having 0.12ms latency. This tells us there is about 0.05ms latency added to the equation when you use the HDDBoost system but that is a performance hit we are willing to take any day of the week as it is so small as to not only be unnoticeable but negligible in the grand scheme of things.

 

[AkG]

XP Start Up / Adobe CS3 Load Time

XP Start Up

HDDBoost = WD 320GD HDD + Kingston SSDNow 40GB

When it comes to hard drive performance there is one area that even the most oblivious user notices: how long it takes to load the Operating System. For our tests the clock starts as soon as the system “beeps” and stops when our Anti-Virus splash screen disappears. While all the other tests were run with a streamlined XP image this particular image is the test bed's “day to day” OS and it has accumulated a lot of crud over the months from installs and removals. We chose the Anti-Virus splash screen as our finish line as it is the last program to be loaded on start up.

Note that we ran the tests with and without Sync done to show if the HDDBoost itself adds any unnecessary overhead

​

The first test here really does look pretty good. Basically, the addition of an inexpensive SSD allows an older, extremely fragmented hard drive to perform nearly at the level as one of the fastest spindle-based drives on the market.

Adobe CS3 Load Time

HDDBoost = WD 320GD HDD + Kingston SSDNow 40GB

Photoshop is a notoriously slow loading program under the best of circumstances, but when you add in a bunch of extra brushes and thesuch you get a really great torture test which can bring even the best of the best to their knees. Let’s see how our review unit faired in the Adobe crucible!

​

Once again we can see that this amazing little piece of technology can take a below average hard drive and turn it into a real winner.

 

[AkG]

Real World Data Transfers

Real World Data Transfers

HDDBoost = WD 320GD HDD + Kingston SSDNow 40GB

No matter how good a synthetic benchmark like IOMeter or PCMark is, it cannot really tell you how your hard drive will perform in “real world” situations. All of us here at Hardware Canucks strive to give you the best, most complete picture of a review item’s true capabilities and to this end we will be running timed data transfers to give you a general idea of how its performance relates to real life use. To help replicate worse case scenarios we will transfer a 4.00GB contiguous RAR file and a folder containing 49 subfolders with a total 2108 files varying in length from 20mb to 1kb (1.00 GB total).

Testing will include transfer to and transferring from the devices, timing each process individually to provide an approximate Read and Write performance. To then stress the dive even more we will then make a copy of the large file to another portion of the same drive and then repeat the process with the small one. This will test the drive to its limits as it will be reading and writing simultaneously. Here is what we found.

<img src="http://images.hardwarecanucks.com/image/akg/Storage/HDDBoost/copy_lg.jpg" border="0" alt="" />

<img src="http://images.hardwarecanucks.com/image/akg/Storage/HDDBoost/copy_sm.jpg" border="0" alt="" />

<img src="http://images.hardwarecanucks.com/image/akg/Storage/HDDBoost/copy_self.jpg" border="0" alt="" />
​

After running these tests we can say with reasonable certainty that the device does not swap out data on the solid state drive unless you explicitly tell it to via the HDDBoost Utility or by rebooting the system. This to us is a good thing and these numbers really do show that while you may not get the full power of the solid state drive, you do get most of it. In return for a little bit of lost power you do get the freedom of not needing to worry about firmware updates, AHCI vs. IDE, ITGC, TRIM or other terms which may be gobbledygook to the average consumer. That is a trade we are sure many people will be willing to make.