by Andrey Kuznetcov
01/06/2007 | 11:24 AM
Western Digital’s external hard disk drives aren’t frequent guests in our labs, so this is a special occasion for two reasons at once. First, we’ve got a new drive from WD which is called MyBook Pro Edition. And second, this hard disk drive is equipped with three interfaces to connect to the PC.
<%BANNER[article]%>We have often seen a drive’s performance being limited by USB 2.0 and FireWire 400 just because HDDs have progressed faster than the interfaces. You can put it another way: the interfaces have fallen behind.
Moreover, the storage capacity of HDDs is growing up at a tremendous rate. Considering the trend to create multi-disk external drives (e.g. the Bigger Disk Extreme from LaCie or the Maxtor OneTouch III Turbo Edition, whichw e discussed in our article called Maxtor OneTouch III Turbo Edition Hard Disk Drive with 1TB Storage Capacity), the requirements to the interface bandwidth have grown manifold!
Using the FireWire 800 interface is one way to solve the problem. We’ve had enough opportunities to see how efficient it is, but it is not widespread and few manufacturers implement it in their products. We’ve got a kind of vicious circle here: they don’t make HDDs because there are few PCs with that interface, but PCs don’t support the interface because there are few peripheral devices with it. If it were not for Apple with its progressive ideas (“don’t follow the market, but lead it where it’s good for you”), we’d have to do with USB 2.0 even today.
This situation is changing, though. The external eSATA interface is becoming popular and offers high bandwidth and multiplexing (from a user’s point of view, this term means the option of connecting several devices to a single port). But FireWire 800 is a good interface anyway and is also backwards-compatible with camcorders and other photo/video equipment, which may come in handy for a home PC.
But now let’s take a look at the new HDD from Western Digital.
This HDD series has a most appropriate name: the drive we’ve got for our tests looks like a thick book indeed. Its case is made of thick silvery plastic. There are a lot of vent holes in the side panels, except for the “binding edge”, that are to ensure proper thermal conditions for the HDD. You can position the device both vertically or horizontally for work. It takes less space when placed upright. There’s a band of a rubber-like plastic going along the perimeter of this “book” so that it stood firm on any surface. The Power button is inserted into two LED rings on the “binding edge”: the external ring indicates power and disk activity and the other shows how much data the drive currently stores. On the opposite side of the case there are two FireWire 800, one FireWire 400 and one USB 2.0 port, a power connector, and a Kensington lock hole.
The case contains a 3.5” hard disk drive with a storage capacity of 500GB. Its spindle rotation speed is 7200rpm and its cache buffer is 16MB large. The average seek time is 8.9 milliseconds. The drive supports three interfaces (FireWire 800 and 400, and USB 2.0), Automatic Power On and Safe Power Off features. The drive is operable within an ambient temperature range of 5° to 35°C. Its dimensions are 172.2 x 143 x 56.7mm. Its weight is 1.2kg.
The following is included with the drive: an external power adapter with cable, FireWire 800 cable, FireWire 400 cable, USB 2.0 cable, Quick Installation Guide, and a CD with the automatic backup utility EMC Retrospect Express.
The average retail price of the drive is $350.
The following testing utilities were used to explore the operating properties of the external hard drive:
Testbed configuration:
The external drive was connected to a mainboard’s USB 2.0 port during the test. The FireWire 400 interface was provided by the VIA VT6307 chip installed on the mainboard.
The drive’s results will be compared with those of products with a similar storage capacity. We took a Seagate Pushbutton Backup ST3750640CB-RK and a Maxtor OneTouch III FireWire 400/USB 2.0 F01G500 (the detailed review of Seagate's solution will follow shortly). These two drives do not support FireWire 800.
I’ll first run a few synthetic tests with the help of IOMeter. In the sequential read/write test the drive is receiving a stream of read/write requests with a request queue depth of 4. The size of the requested data block is changed once every minute. Thus, this test shows the dependence of the drive’s sequential read/write speed on the size of the requested data block. For the sake of readability, shortened names for the drives and abbreviations for the interfaces will be used here and elsewhere in the review.
Sequential reading comes first. The diagram shows the advantage the Western Digital drive is given by its support of the FireWire interface: the max sequential read speed is much higher in this case – almost two times the speed the drive has when using USB 2.0. The WDG1T5000N also performs faster with FireWire 400 than with USB 2.0. Comparing the drive to its opponents, they are all roughly similar when using USB 2.0. With FireWire 400, the Western Digital is slower than the Maxtor OneTouch III. It is also slower with FireWire 800 when reading small data blocks (up to 4KB).
The sequential writing diagram proves the highest efficiency of FireWire 800, too. Using this interface, the Western Digital delivers two times its own performance with FireWire 400. But it shows the slowest speed when connected via USB 2.0.
As for its standing among the opponents, the WDG1T5000N is slower at sequential writing than the Maxtor OneTouch III when both are using FireWire 400. The same is true for FireWire 800, but only when the WDG1T5000N is processing small data blocks (2KB and smaller). When the drives are all connected to USB 2.0 ports, the Western Digital is the slowest of them.
The Workstation pattern simulates the typical load on a workstation at request queue depths up to 32. I performed this test on the full capacity of the drives as well as on 32GB partitions created on them.
To calculate the overall performance rating of a drive in this test, we use the following formula:
Performance = Total I/O (queue=1)/1 + Total I/O (queue=2)/2 + Total I/O (queue=4)/4 + Total I/O (queue=8)/8 + Total I/O (queue=16)/16 + Total I/O (queue=32)/32.
The Western Digital is superior in this test. It delivers the highest overall performance with both USB 2.0 and FireWire 400. Note that the FireWire 800 interface doesn’t improve the overall performance of the WDG1T5000N much.
These patterns simulate the load typical of the disk subsystems of File and Web servers. The request queue depth is varied from 1 to 256 outstanding requests. The drive’s performance rating is calculated in this test by averaging its speed under all possible loads.
Here are the drives’ ratings in the File Server pattern:
The WDG1T5000N enjoys a complete victory over its opponents irrespective of the interface employed. Note also that the type of the interface has a small effect on the efficiency of the drives.
The results of the Web-Server test are similar to the previous one. The drive from Western Digital is beyond competition and the type of the interface doesn’t influence its performance as well as that of its opponents much.
In the Database pattern the drive is processing a stream of requests to read and write 8KB random-address data blocks. The ratio of read to write requests is changing throughout the test. The request queue depth is varied from 1 to 256 requests.
I limited myself to building only three diagrams out of these numbers:
When the queue has the shortest possible length, the Western Digital drive is faster in almost every case except for write-only mode. The efficiency of the WDG1T5000N doesn’t depend on the interface it uses (FireWire 400 even looks preferable to FireWire 800 here).
Increasing the request queue depth to 16 requests doesn’t bring any changes in the standings of the drives. The Western Digital is faster than its opponents, save in write-only mode when it is again slower than the Maxtor OneTouch III. There’s no difference between the interfaces: the use of FireWire 800 doesn’t bring any profit to the tested drive.
The third diagram shows the results of the drives at the maximum queue length. Like in the previous two cases, the Western Digital delivers higher performance than the other two devices, again with the exception of write-only mode where it is somewhat slower than the Maxtor OneTouch III. And again we see no difference between the three interfaces.
The multi-threaded tests simulate a situation when there are several (from one to four) clients accessing the drive at the same time and the drives are processing requests for sequential reading or writing.
So, I use IOMeter to send requests to read/write 64KB data blocks with a steadily increasing address. There are from one to four clients (Workers in IOMeter’s terminology). Each client processes its own disk space range (there is a difference of 8GB between the ranges). The depth of the outgoing request queue is varied from 1 to 8.
I built diagrams for the case with the shortest request queue. It is the most common situation in practice.
So, multi-threaded reading comes first.
The Western Digital is generally slower than its opponents when working via USB 2.0 and FireWire 400. But when connected to a FireWire 800 port, this drive leaves the others behind, especially when there is only one hypothetical user. The advantage is smaller when there are more clients accessing the drive.
Now, let’s see what we have at multi-threaded writing.
It’s somewhat different than in the previous test: the Western Digital is the slowest with USB 2.0 and second with FireWire 400. When connected via FireWire 800, the WDG1T5000N leaves no chance to its opponents at any load – it knows no rivals when it comes to multi-threaded writing.
I ran WinBench 99 to create data-transfer diagrams for the Western Digital drive. These are indicative of the effective bandwidth of each of its interfaces.
USB 2.0
FireWire 400
FireWire 800
These diagrams prove the high efficiency of the FireWire 800 interface. It allows the WDG1T5000N to show its full speed – the graph starts at a point of about 75MB/s! The diagrams for FireWire 400 and USB 2.0 show a nearly straight line, which means that the interface limits the drive’s speed. This is especially conspicuous with USB 2.0.
Now I will check the real-life performance of the hard disk drives with the FC-Test utility which measures the time it takes to create (i.e. write), read and copy certain file-sets which differ in the number and size of the files included, and then calculates the speed of the drive. This utility gives you an objective picture of performance because it emulates the real operating situations for the hard drive.
The Windows and Programs patterns consist of many small-size files, while the MP3, ISO and Install patterns include fewer files of a bigger size. Two 32GB partitions are created on the drives for the copy operations, and the file-sets are then copied within one partition (Copy Near) and from one partition to another (Copy Far). The creation (writing) and reading of the file-sets is done in the first partition.
The drives will be first tested with FAT32.
The first diagram shows the speed of writing (creating) the file-sets. The Western Digital finds itself second after the Maxtor OneTouch III when using USB 2.0 or FireWire 400. But when connected via FireWire 800, the hero of this review is far faster than its opponents.
The WDG1T5000N can’t compete with the other devices when they are all using USB 2.0. When connected via FireWire 400, it takes second place after the Maxtor OneTouch III. But it is the FireWire 800 interface that allows the Western Digital drive to show its full strength. It enjoys a great advantage over its opponents then.
Copying files within the same partition, the WDG1T5000N is third with large files and second with small files – with the USB 2.0 interface. With FireWire 400 it is second after the Maxtor OneTouch III. FireWire 800 makes the Western Digital drive an unrivalled leader of our races.
The results of this test (copying files from one partition to another) are similar to those of the previous one. When connected via USB 2.0, the WDG1T5000N is second in two patterns that consist of a large number of small files and third in the other three patterns. When connected via FireWire 400, it is overall faster than the Seagate, but slower than the Maxtor OneTouch. But when you connect it to a FireWire 800 port, the drive leaves its opponents far behind.
Now let’s see what we have in the newer NTFS file system.
Using USB 2.0, the WDG1T5000N is only slower than the Maxtor OneTouch at writing (creating) files. With FireWire 400 it is the fastest in two small-files patterns and second with the other three file-sets. FireWire 800 puts the Western Digital drive high above its opponents.
When it comes to reading the file-sets, the WDG1T5000N is somewhat slower than its opponents with USB 2.0 and becomes second with FireWire 400. And then it again beats its competitors when connected via FireWire 800.
Copying files within the same partition puts the Western Digital drive second with FireWire 400 and last with USB 2.0. As usual, FireWire 800 comes to rescue – using it, the WDG1T5000N meets no competition.
The last diagram shows the speed of copying files from one partition to another. There’s nothing particularly new here: the WD drive is the slowest when working via USB 2.0 and second with FireWire 400. FireWire 800 gives it a great boost and it leaves its competitors behind with ease.
This review has shown that any jewel needs a proper setting. However fast a modern external hard disk drive may be, its speed is going to be limited by the widespread interfaces USB 2.0 and FireWire 400. Our tests of the Western Digital WDG1T5000N prove that point.
Using the mentioned interfaces, it has been roughly equal to its opponents, similar products from Maxtor and Seagate (I guess I should be talking about them as about one firm now).
It’s only when you connect the drive via FireWire 800 that it provides a much higher speed. So, I recommend the WDG1T5000N for every user whose computer supports FireWire 800. This will give you a highest practical speed. If you are going to connect your external HDD via other interfaces, you may as well choose any other product (well, you may also be enticed by the stylish design of the WD drive).
I hope the manufacturers will be producing more external HDDs with support of FireWire 800 in addition to other interfaces because HDDs just cry for that – this interface will allow them to show their full speed potential.
The eSATA interface is currently the only real opponent to FireWire 800, but this opposition will make the topic of another review.