by Aleksey Meyev
02/16/2009 | 06:30 PM
While there is a pause in the manufacturers’ race for the highest-capacity hard disk, we have accumulated an almost complete collection of 2.5-inch HDDs with a capacity of 320GB and a spindle rotation speed of 5400rpm. There are six models in total, one from each manufacturer. We guess it is a good reason to perform a comparative test session. 320GB drives have become quite ordinary by now. This storage capacity is not the highest available today, because 500GB models have entered the market, but 320GB drives are nonetheless interesting. Their capacity is quite enough for a majority of notebook users while their speed characteristics are impressive because early 500GB drives reached their capacity by increasing the number of platters rather than through an increase in recording density. Every HDD in this review has two 160GB platters whereas 500GB drives have three 167GB platters, so there is almost no growth in recording density. Our comparative review of 320GB and 500GB drives proved that point (as a matter of fact, a 320GB drive was the fastest in that test session).
Now let’s meet the newcomers.
MHZ series drives are the latest generation of HDDs from Fujitsu. Depending on the second part of the series name, these can be three-platter drives with a spindle rotation speed of 4200rpm (MHZ 2BT), popular 5400rpm models with one or two platters (MHZ 2BH), quick 7200rpm versions (MHZ 2BJ), and 7200rpm drives with an integrated data encryption module using a 256-bit AES algorithm (MHZ 2CJ). We’ve got a two-platter MHZ 2BH with a spindle rotation speed of 5400rpm and a data density of 160GB per platter. The sub-series is rather short: our drive is the largest in it while the smallest one has a capacity of only 80GB (a 40GB model used to sell for while, too). The HDD has standard specs for its class, including 8 megabytes of cache memory.
Life plays odd tricks sometimes. For example, the 320GB model of Hitachi’s latest generation of HDDs has reached us long after the 500GB model (which took part in the above-mentioned comparison of 320GB and 500GB drives).
Like Fujitsu, Hitachi offers a few families of 2.5-inch drives with 350GB models. Besides the standard 5400rpm model we’ve got, there is a 7200rpm model (7K320 series), three-platter models of the 5K500 series, and two nonstandard 5400rpm variants. One of them, the so-called EA model, has the letter E instead of S as the third symbol in the name and, according to the manufacturer, suits best for continuous all-day operation. The company doesn’t say what exactly is changed in this drive to make it more suitable for such load, though. And another is a drive with hardware encryption (its name ends in 01 rather than 00). Contrary to Fujitsu’s products, it is 5400rpm rather than 7200rpm drives that feature such encryption in Hitachi’s line-up. Their interface bandwidth is also limited to 1.5Gbps as opposed to the other models’ 3Gbps.
Again, we will test the standard model that doesn’t have any improvements or special features.
The last new drive comes from Seagate. Like the other makers, Seagate offers a wide variety of models: with higher spindle rotation speed (7200.3) and hardware encryption (5400 FDE.3 and 7200 FDE). We are going to deal with a model from the basic series (5400.5) which has a standard 8MB cache buffer.
The other drives in this test session were already covered in our reviews. These are the first 320GB drives we laid our hands on, Toshiba MKxx52GSX and Western Digital Scorpio Blue, as well as the Samsung SpinPoint M6S that came to us later.
The drives to be tested share the same specifications, except for the model from Western Digital which has twice more cache memory than the others. The next table shows the firmware versions of the drives:
The following testing utilities were used:
We installed the generic OS drivers for the drives and formatted them in FAT32 and NTFS as one partition with the default cluster size. For some tests 32GB partitions were created on the drives and formatted in FAT32 and NTFS with the default cluster size, too. The drives were connected to a Promise SATA300 TX4302 controller installed into a PCI-X slot and switched from the quiet mode (with Advanced Acoustic Management enabled) into the ordinary operating mode where necessary.
We are going to get rid of obsolete benchmarking suites, so we recorded the data-transfer graphs of the drives with our own IOMark tool rather than with WinBench:
The following diagram compares the read speed at the beginning and end of the partition created on each drive:
The Samsung seems to have the highest recording density of all the drives, but that’s not so. If you look at the graphs, you can see that some of them have small slumps at the beginning, which is reflected in the diagram. The speed at the end of the partition is the true indication of what drives have higher recording density: these are the HDDs from Samsung, Seagate and Western Digital.
IOMeter is sending a stream of read and write requests with a request queue depth of 4. The size of the requested data block is changed each minute, so that we could see the dependence of the drive’s sequential read/write speed on the size of the data block. This test is indicative of the highest speed the drive can achieve.
The numeric data can be viewed in tables. We’ll discuss graphs and diagrams.
IOMeter: Sequential Read results
The three leaders of the previous test confirm their results and deliver somewhat higher maximum sequential speeds than the other drives. The Samsung boasts the highest top speed at very large data chunks but it is the slowest with small data chunks. The Toshiba is effective at gluing the requests together and has the highest speed with small data blocks, but is no record-breaker with large ones. The Seagate seems to be the best overall as it shows high efficiency irrespective of the size of the data block. The HDD from Western Digital is also good overall, but its results are somewhat lower.
IOMeter: Sequential Write results
This time the HDD from Western Digital is the overall leader despite the problem with 4KB data chunks (this problem doesn’t show up in practical applications, but only in the synthetic benchmark). The Seagate also has good results, though. Fujitsu’s performance with small data blocks and the surprisingly low speed of the Toshiba and Samsung are the disappointments of this test.
In this test IOMeter is sending a stream of requests to read and write 512-byte data blocks with a request queue of 1 for 10 minutes. The total number of requests processed by the HDD is over 60 thousand, so we get a sustained response time that doesn’t depend on the HDD’s buffer size.
We’ve got predictable and close results from five HDDs, among which the Western Digital is the best. The Seagate, the sixth drive, is more than 4 milliseconds worse than the others. This is a considerable gap which is sure to affect the drive’s performance at random operations. This doesn’t look like a quiet operation mode – the heads positioning unit of this drive just seems to be slower than the ones of the other HDDs.
However, the Seagate is even worse at writing. Its write response is just terrible. Some time ago products of this brand used to have such results, too. This actually meant the lack of deferred writing. Could Seagate have returned to this far-from-good trait of its older firmware?
As for the other HDDs, we can note the good results of the Fujitsu and Western Digital. They seem to be doing deferred writing just fine.
Now we’ll see the dependence between the drives’ performance in random read and write modes on the size of the data block size.
We will discuss the results of the disk subsystems at processing random-address data in two versions. For small-size data chunks we will draw graphs showing the dependence of the amount of operations per second on the data chunk size. For large chunks we will compare performance depending on data-transfer rate in megabytes per second. This approach helps us evaluate the disk subsystem’s performance in two typical scenarios: working with small data chunks is typical for databases. The amount of operations per second is more important than sheer speed then. Working with large data blocks is nearly the same as working with small files, and the traditional measurement of speed in megabytes per second becomes more relevant.
IOMeter: Random Read, operations per second
It is the response time of the drive that is the most important factor when reading in small data chunks. Therefore we’ve got five drives going close to each other, and one slow Seagate. First place goes to Western Digital, but its advantage is negligible.
IOMeter: Random Read, MBps
The Samsung takes the lead when the data chunks are larger. The Seagate doesn’t lose too much here because the speed of sequential operations becomes a factor, too. As a result, the HDDs are overall very close to each other in this test.
IOMeter: Random Write, operations per second
Like at reading, writing in small data blocks depends on the drive’s response time. But as opposed to reading, the resulting speed depends not on how fast the heads are moving around the platters and not on the recording density per platter but on the firmware algorithms, particularly on the depth of deferred writing and on the efficiency of requests reordering. We’ve got two clear leaders here: the drives from Western Digital and Fujitsu. The Seagate is the loser. The speed of the latter drive doesn’t depend much on the size of the data chunk. It is not going to show us any deferred writing at all in IOMeter Database.
IOMeter: Random Write, MBps
It is the sequential speeds of a drive that affect its performance here. As a result, the Western Digital wins this test while the Seagate gets closer to the others.
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 from 0% to 100% with a step of 10% throughout the test while the request queue depth varies from 1 to 256.
You can click the following link to view the tabled results in IOMeter: Database.
We will build diagrams for request queue depths of 1, 16 and 256. There are too many results, so we divide them into two diagrams here and in some of the later tests.
Alas, our apprehensions come true: the Seagate does not have deferred writing at all. And it is not really fast at reading, either. It is interesting to compare the leaders: the Western Digital accelerates steadily as the percentage of writes in the queue is increasing whereas the Fujitsu wakes up rather late but speeds up fast, outperforming the Western Digital at pure writing. Anyway, we like the behavior of the HDD from Western Digital more because it has the best results under the most complex loads (when the hard disk is receiving about the same share of reads and writes).
Request reordering begins to work when we increase the load. This saves the day for the Seagate when the percentage of reads is high (this HDD has efficient NCQ) but it has no deferred writing anyway. The HDD from Western Digital is the obvious leader still. The Fujitsu might challenge it, but had a sudden slump at 10% writes. Its firmware is not ideal, either.
When the request queue is very long, the driver begins to affect the performance. Thanks to it, the Seagate speeds up at low percentages of writes and even shows something like deferred writing. It is far slower than the other HDDs, though. The Western Digital is brilliant again while the Fujitsu is second best.
And now for each of the three new drives we will publish a diagram with five graphs for five request queue depths. This helps better understand the behavior of a HDD and the specifics of its firmware algorithms.
Comparing this diagram with the one of Fujitsu’s previous series, MHY 2BH, we can see the progress in this company’s firmware. The efficient deferred writing is now combined with efficient request reordering algorithms. If it were not for the slumps at 10% writes, the firmware would be perfect.
The Hitachi drive is consistent under any load, showing good scalability and medium-efficiency deferred writing. It is good that there are no slumps in the middle of the graphs, but NCQ might be more effective.
We have said all we think about the Seagate drive above, so we can only repeat our point. This model is very slow at reading and does not support deferred writing. Its efficient NCQ algorithms cannot help it much.
You can view such diagrams for the other participated HDDs in the above-mentioned reviews.
The drives are tested under loads typical of servers and workstations.
The names of the patterns are self-explanatory. The Workstation pattern is used with the full capacity of the drive as well as with a 32GB partition. The request queue is limited to 32 requests in the Workstation pattern.
The results are presented as performance ratings. For the File-Server and Web-Server patterns the performance rating is the average speed of the drive under every load. For the Workstation pattern we use the following formula:
Rating (Workstation) = 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.
First place goes to Western Digital here. It is predictably followed by the Fujitsu whereas the Seagate is rather surprisingly third. The lowest performance in this test is delivered by the Toshiba whose firmware obviously has some problems that show up as the odd irregular dependence of performance on the request queue depth.
The final ratings just prove what we have already seen well enough in the diagram.
The standings change a lot with the new load which now has write requests. For example, the HDD from Western Digital is now far ahead of the other HDDs irrespective of the request queue depth. The Fujitsu and Toshiba share second place whereas the Seagate is an outsider, being much slower than the leader at short queue depths.
The performance ratings indicate the huge advantage of the Western Digital drive over the others.
Western Digital is not the leader anymore when it comes to the workstation load. It is outperformed by the Fujitsu at each but the shortest request queue depth. The Samsung and Toshiba keep close behind the leaders whereas the Hitachi is noticeably slower. The imperfect firmware algorithms of the Seagate make it far slower than the other drives. It is just out of competition here.
Five HDDs have similar performance ratings in this test whereas the Seagate is much worse than the others. Since our rating formula gives higher weights to shorter request queue depths, the Western Digital is somewhat better than the Fujitsu while the Toshiba is as good as the latter.
When the test zone is limited to 32 gigabytes (or 10% of the total capacity), the HDDs all deliver similar results, the Fujitsu being a little better than the others. The Western Digital is somewhat slower than the others at short queue depths for some reason.
Save for the Western Digital, the drives have similar performance ratings in this test.
The multithreaded tests simulate a situation when there are one to four clients accessing the virtual disk at the same time – the clients’ address zones do not overlap. We’ll discuss diagrams for a request queue of 1 as the most illustrative ones. When the queue is 2 or more requests long, the speed doesn’t depend much on the number of applications. You can also click the following links for the full results:
The Samsung wins when reading one thread, but slows down when there are more threads to process. However, it is the Toshiba drive that is the slowest of all: its speed drops fourfold. The drives from Seagate and Western Digital are the best at multithreaded reading. The Fujitsu behaves oddly: it loses its speed greatly at two threads but accelerates somewhat at three threads whereas every other drive slows down then.
It’s simple with multithreaded writing: we can clearly see both leaders and losers. The Samsung copes best with this test. The Fujitsu and Western Digital are good, too. The Toshiba and Hitachi are the slowest here, suffering a 50% performance hit with multiple write threads.
For this test two 32GB partitions are created on the disk and formatted in NTFS and then in FAT32. After that a file-set is created. It is then read from the disk, copied within the same partition and then copied into another partition. The time taken to perform these operations is measured and the speed of the disk is calculated. The Windows and Programs file-sets consist of a large number of small files whereas the other three patterns (ISO, MP3, and Install) include a few large files each.
We’d like to note that the copying test is indicative of the drive’s behavior under complex load. In fact, the HDD is working with two asynchronous threads (one for reading and one for writing) when copying files.
This test produces too much data, so we will only discuss the results achieved with the Install, ISO and Programs file-sets in NTFS. You can use the links below to view the full results:
Western Digital’s model wins this test, being the fastest drive to write every file-set. The Samsung might challenge the winner but its performance proved to depend heavily on the size of the processed files. It didn’t like the mixed set of large files from the Install pattern. The other drives are similar to each other with the exception of the Seagate which falls behind with large files.
The Samsung is in the lead when reading files. It now copes well with every file-set. The HDD from Western Digital looks good, too, handling small files very efficiently (it did not get to be the leader only because it didn’t win with large files). The Seagate is a disappointment: being very fast with larger files, it is the worst drive of all with small ones. The Hitachi is not fast, either.
The Samsung is the best copying files within the same partition. The HDD from Western Digital overtakes it in the Install pattern only, and by a negligible margin. The other HDDs go close to each other, excepting the Toshiba which is the slowest with every file-set and the Hitachi which is surprisingly fast at copying large files.
The standings do not change much when we increase the distance of copying. The HDD from Western Digital now wins with the smallest files whereas the Samsung is ahead with the other two file-sets. The Toshiba has left its last place to the Seagate. Both are very slow, though.
PCMark04 benchmarks drives in four different modes: Windows XP Startup is the typical disk subsystem load at system startup; Application Loading is the disk activity at sequential starting-up and closing of six popular applications; File Copying measures the HDD performance when copying a set of files; the General Usage parameter reflects the disk activity in a number of popular applications. These four parameters are used to calculate the overall performance rating.
We ran each test ten times and averaged the results.
The Fujitsu copes best with launching the OS. The HDDs from Western Digital and Hitachi share second place while the Toshiba is the loser, again.
The Fujitsu is in the lead in the Application Loading subtest, too. Second place goes to Western Digital, and third place goes to Hitachi. The Seagate is the slowest drive here.
Interestingly, unlike in many other test sessions, the File Copying results do not agree with the results of our FC-Test. The HDD from Western Digital is in the lead, followed by the Fujitsu. The Samsung was brilliant in FC-Test but takes only fourth place here. The Toshiba is the loser.
The Fujitsu wins this subtest, too. The WD drive, an obvious favorite of this test session, has second place. The drives from Samsung and Seagate are considerably slower than the others here.
Summing up the results of this test, we’ve got three leaders, the drives from Fujitsu, Western Digital and Hitachi, and the slower other drives. First place goes to Fujitsu whereas the Seagate and Toshiba share last place.
PCMark05 is an updated version of the previous benchmark. Instead of File Copying, there is now a File Write trace. A new trace called Virus Scan is added. Its name is self-explanatory.
Again, we performed each test ten times and averaged the results.
The new version of the benchmark puts the Fujitsu on top, too. The Toshiba is the last drive again.
There is nothing new in the Application Loading subtest. We’ve got the same leaders and losers.
The third subtest produces the same result as the previous version of the benchmark. So, we guess it’s time for us to dump one of our PCMarks.
Here is the new Virus Scan subtest. There are two leaders here: the Fujitsu (this drive feels at ease in this group of tests) and Seagate. The HDD from Western Digital sinks to last place, falling behind the three medium-speed drives.
The File Write test is new in this version of the benchmark. The HDD from Western Digital has been better than the others at writing throughout this test session and wins this test, too. The Seagate takes second place to our surprise. It is even more surprising to see the Samsung in last place.
The final standings are somewhat different, but the two top places still go to Fujitsu and Western Digital. The Seagate is third now, however. The Samsung is the loser in this version of PCMark.
To make this part of our test session complete, we are going to run the latest version of PCMark called Vantage. Compared with the previous versions, the benchmark has become more up-to-date and advanced in its selection of subtests as well as Windows Vista orientation. Each subtest is run ten times and the results of the ten runs are averaged.
Here is a brief description of each subtest:
Basing on these subtests, the drive’s overall performance rating is calculated.
We have already got used to having the drive from Western Digital among the leaders but it is somewhat unusual to see the Seagate take first place. Its firmware may be no so bad after all. Besides high speed of sequential operations, it boasts optimizations for certain common types of load. The Toshiba and Samsung are on the losing side here.
Our tests change but the losers do not. The winners are the same, too. The drives from Western Digital and Fujitsu are the best in the Gaming test.
Photographers will probably prefer other drives, however, because the HDD from Western Digital sinks to last place in the Photo Gallery test. The Seagate is first, followed by the Fujitsu.
This time PCMark emulates Windows Vista, and there are no clear outsiders now. However, we’ve got a couple of drives that are somewhat faster than the others: these are the drives from Western Digital and Fujitsu.
The Movie Maker test seems to be a very specific load: the inconspicuous Hitachi manages to take second place, closely behind the leading Western Digital. This load is disliked by the fastidious Samsung as well as by the Fujitsu.
Once again we have to admit the odd peculiarity of this test. File caching plays an important role here, allowing some HDDs to pass the test at a speed higher than its sequential read speed. The winner Samsung is indicative of that. The second-best Seagate nearly achieves its sequential read speed, too. The Toshiba has its traditional last place again.
The Fujitsu copes best with adding files into the player. Having been an outsider throughout this test session, the Toshiba is as good as second here. The Seagate is the loser in this test.
The Application Loading test now contains new applications, which affects the HDDs’ standings. The Toshiba is first (it has waken up too late, though – the review is about to end soon), and the Samsung is the worst.
There are more subtests in the new benchmark and they are very different from the earlier versions’, but the HDDs from Fujitsu and Western Digital still have the highest overall ratings. The Samsung performed poorly here.
You can refer to our article called Hard Disk Drive Power Consumption Measurements: X-bit’s Methodology Indepth for details on this test. We’ll just list the specific modes we measure the power consumption in:
Let’s check out each mode one by one.
The HDD from Western Digital requires the highest current to start up. It is the only drive to need more than 1 ampere. This means even two USB ports simultaneously won’t be enough to power it up in a mobile rack (to remind you, one USB port guarantees a current of no more than 0.5 amperes; any load above this value is beyond the requirements of the USB specification). The drive seems to require such a high current to switch into operation-ready mode in the shortest time possible. That’s good, but we guess the developer should have stopped at 1 ampere.
The Samsung is the most economical drive in the test, requiring only two thirds of the current demanded by the Western Digital. The Hitachi needs somewhat more, taking second place.
Most of the drives need somewhat less than 1 watt in idle mode. The good exceptions are the Toshiba and the Fujitsu.
It is the electromechanical section of a HDD that consumes most power during random reading whereas the electronics is mostly idle. Here, the Seagate is by far the most economical drive. Recalling its poor results at random reading (and its high response time), we can suspect that its heads unit has a weaker but more economical actuator. The leader is followed by the Samsung and Toshiba that have similar results. The HDD from Western Digital is the most voracious drive in this test.
When a hard drive is performing random writing, its electronics consumes more and the overall power consumption grows up. The standings differ somewhat because the growth is different with each HDD. The Seagate is still the most economical model (we can also recall that it showed no deferred writing in IOMeter Database and had low results at random writing) while the Fujitsu and Hitachi share the title of the most voracious model. The low growth of power consumption of the Western Digital drive is somewhat surprising: its electronics either consumed a lot at reading or is very power efficient.
The difference between reading and writing is smaller when it comes to sequential operations. The Seagate is the most economical again whereas the drive from Western Digital is still very hungry, asking for 3 watts of power. That’s quite a lot and exceeds the capability of one USB port. You should keep this fact in mind if you are choosing a hard disk to use in a mobile rack. The high speed of the HDD from Western Digital is accompanied with high power consumption.
We’ll try to summarize the results of this test session now.
The best models come first. These are the Western Digital Scorpio Blue, which had been among the first to arrive to our test lab and still maintains its superiority across a majority of our tests, and the Fujitsu MHZ 2BH. The latter is a nice discovery for us. Fujitsu not only increased recording density per platter in this series but also revised the firmware. As the result, the effective writing is now combined in this HDD with advanced request reordering algorithms. This HDD was a leader through most of our tests, successfully competing with the Scorpio Blue. It is hard to tell the best one of these two because the Western Digital is somewhat better in the server tests and at writing whereas the Fujitsu is faster in PCMark and has lower power consumption.
The Hitachi Travelstar 5K320, Samsung M6S and Seagate Barracuda 5400.5 are each good from a specific point of view only. The Hitachi is free from obvious drawbacks, delivering medium performance throughout this test session. The Samsung was excellent in FC-Test and at multithreaded writing, but was far from brilliant in the other tests, rising up to top place but occasionally. The same goes for the Seagate. It did well at multithreaded reading and in some of PCMark tests, but was uncompetitive to the other tested HDDs under server loads due to its terribly high response time and lack of deferred writing. The good thing about this model is its lowest power consumption which may be a significant advantage in some applications (but we don’t think it justifies the performance hit if you compare this drive’s results to those of the Fujitsu, for example).
And finally, the Toshiba MKxx52GSX is not exceptional at all, save for being regularly in the bottom position in the diagrams. Toshiba should take care to turn out better drives the next time they make them. Otherwise, they will have to compete for the customer by means of the price factor only, which may be no good for the company’s budget.
We are now waiting for newer, denser and larger disks since there are all the prerequisites for that. And we also hope to get some 7200rpm models which currently seem to be somewhat immaterial and only exist on the Web and in the manufacturers’ press releases.