by Aleksey Meyev
08/05/2008 | 05:43 PM
The market of hard disk drives seems to be the most difficult one for PC hardware makers when it comes to fighting for the customer. Distinguishing one’s brand with an eye-catching exterior design of the product is not an option. Western Digital was the only manufacturer who tried to do something like that with the version of its Raptor drive that had a transparent window. It is also impossible to drop the product price much lower than the competitors’ prices: the profitability is already low due to the tough competition whereas every company also has to do some research and improve the manufacturing process. Attracting the customer with exclusive technologies is not an option, either. Most of the technologies are utilized by every manufacturer (for example, all of them now park the heads on the ramp and use perpendicular recording) or the point of a technology may be so complex that it wouldn’t be possible to explain it in easy language to the common user. Thus, the only characteristic which is clear to everyone is storage capacity and every manufacturer is trying to increase it more and more.
The race for higher storage capacities is not something useless, though. Large applications with their data, high-definition video content, collections of music and photos all need to be stored somewhere. The hard disk is often the first component that asks for an upgrade in a new computer. While the system may still deliver enough performance for most of your applications, you find yourself having run short of disk space. You can easily solve this problem with a desktop PC by adding a second hard disk, but with a notebook you have to replace the old HDD with a new and larger one.
Large storage capacity also means high speed. This correlation can be explained easily: large disks have platters with higher recording density, which is one of the few available methods of increasing performance. Another method of increasing the drive’s capacity is installing multiple platters into it. This method does not increase the drive’s sequential read speed but lowers its access time because there is a higher chance of one of the heads being near the desired sector.
In our previous review of 2.5-inch drives we discussed models with capacities of 250 and 320 gigabytes. Today we invite you to the next level in order to compare the 320GB models with the Hitachi 5K500 whose storage capacity is 500 gigabytes. Half a terabyte in the 2.5-inch form-factor seemed a fantastic thing just a few years ago. And considering that today’s top-performance notebooks support two drives, we can say that they have passed the 1TB milestone.
The spindle rotation speed of the HDDs we are going to test today is 5400rpm. Such products are a compromise between price and performance. 4200rpm models are rather slow while 7200rpm ones find it harder to reach the same recording density at the current level of technologies (the problem is in accurate reading and writing of information, to be exact) and also have higher power consumption. 7200rpm drives and the 10,000rpm VelociRaptor from Western Digital are targeted at servers, compact desktops and desktop-replacement notebooks whereas mass notebooks need more economical disks. Storage capacity is more important than speed for external storage devices, too.
Now, let’s take a look at the hard drives we are going to benchmark.
Hitachi achieved the storage capacity of 500 gigabytes by installing three 167GB platters into the 5K500, so this drive is just a little ahead of its opponents in terms of recording density. Unfortunately, this solution increased the height of the product by 3 millimeters. It just will not fit into quite a lot of notebooks and external racks. The solution itself is not new. We have already seen three-platter 2.5-inch drives with an increased height of the case. Those were Fujitsu’s models with a spindle rotation speed of 4200rpm. Well, it is quite natural for Hitachi to know that two platters is a conventional but not maximum parameter as this company has mastered the 5-platter design so well in the sector of 3.5-inch models.
The 5K500 has inherited all its technologies from the predecessors. The only innovation is Rotational Vibration Safeguard technology which had been limited to the company’s 3.5-inch drives before. To remind you, the point of this technology is in minimizing vibrations affecting the read/write heads by utilizing velocity sensors. This improves the positioning accuracy of the heads above the necessary track.
It will be interesting to see what performance this record-capacity drive can deliver.
This newcomer is from Samsung’s new SpinPoint M6S series which features 160GB platters just as in the other brands’ products. It is in this series that Samsung has also announced a 500GB model. Like the Hitachi 5K500, it is based on three platters but Samsung claims it to have the standard (9.5mm) height and fit any notebook. It would be interesting to see how they achieved that, but we haven’t got that drive yet. We will test the dual-platter 320GB model instead.
Of course, the new series has inherited all the technologies found in the M5S series and added some more: the higher recording density made it necessary to lower the read/write head closer to the platter surface and control its flight more accurately. As a result, the traditional TMR heads have acquired temperature-based flight control called Fly on Demand.
The 320GB Toshiba MKxx52GSX and Western Digital WD3200BEVT models that we discussed in the previous article are going to take part in this test session, too.
The following table shows the specifications and firmware versions of the hard disk drives:
Click to enlarge
Take note that the Western Digital Scorpio is equipped with 16 megabytes of cache memory. The other HDDs come with an 8MB cache.
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.
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 maximum speed the drive can achieve.
The numeric data can be viewed in tables. We’ll discuss graphs and diagrams.
IOMeter: Sequential Read results
The newcomers are competitive to their opponents at sequential reading. For example, the Hitachi is average if compared with the others whereas the Samsung is slow on small data chunks and fast on large data chunks. It achieves its top speed on larger data blocks than the other drives, though. The Toshiba remains the leader in reading small data blocks – its ability to glue multiple requests into a single large request puts it beyond any competition.
Interestingly, the drives from Hitachi and Western Digital both suffer a performance slump on 4KB data chunks. This is their response to the specific synthetic load when the controller is installed into a PCI-X slot. We will not observe this kind of a slump in the other tests, yet the other HDDs do not have it at all.
IOMeter: Sequential Write results
The standings are totally different at sequential writing: the Hitachi is now competing with the Western Digital for first place while the Samsung shows a very low top speed for such dense platters as it has. This time around it is the HDD from Western Digital that shows its ability to effectively glue multiple small requests into a large one.
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.
The HDD from Western Digital is unrivalled at both reading and writing. It must have a very fast mechanism of the read/write heads backed up by efficient firmware algorithms. The Hitachi is a disappointment: its response time at reading is too high. Even platters with lower recording density were better in this respect.
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
When reading small data chunks, the HDDs are ranked according to their access time. The Western Digital is ahead, and the Hitachi is last, the other two drives being in between.
IOMeter: Random Read, MBps
Access time is not a decisive factor when the drive is reading large data blocks with random addresses. Therefore the Samsung goes ahead thanks to its higher sequential read speed. Its advantage is growing up along with the data block size. The Hitachi and Western Digital have identical results, the Toshiba following them closely.
IOMeter: Random Write, operations per second
The HDD from Western Digital is far faster than the others when writing small data blocks. The other drives go close to each other but the Hitachi slows down and falls behind as the data block size gets larger. Its firmware seems to stumble on such blocks, being unable to pack them fully into the buffer segments.
IOMeter: Random Write, MBps
The HDD from Western Digital is unrivalled when writing large blocks, too. The Hitachi is in last place but its speed is growing up proportionally to the data block size. When the load gets close to sequential writing, the Hitachi overtakes the Samsung and nearly catches up with the Toshiba.
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.
The sprightly Western Digital is ahead of all under minimum load. The Samsung and Toshiba go neck and neck, the former being somewhat faster at reading and the latter, at writing. The Hitachi behaves in a curious manner: it copes well enough with lots of writes but is slow when there is the same share of reads and writes to be performed.
The Toshiba outperforms the Samsung at reading under the increased load. This seems to be due to some flaws in the Samsung’s firmware rather than to the Toshiba’s advantages. Samsung’s drive has an inexplicable performance hit at high percentages of reads. The Hitachi is again good when processing same-type requests but slows down greatly under a mixed stream of requests.
The picture changes at the highest load. This time the Hitachi is very close to the leading Western Digital at pure reading and writing but has modest performance under mixed loads again.
The following diagrams show each drive’s performance at five different queue depths.
The characteristic growth of performance in the area of low percentages of writes indicates that the Hitachi is good at request reordering. This technique shows its full effect under high loads only, though. The drive also has deferred writing, but it seems to conflict with NCQ. These features just don’t work simultaneously. The drive seems to be confused when there is about the same amount of reads and writes – and slows down under such load.
The Samsung M6S is a calm and collected device. It has very discreet deferred writing and does little of request reordering. Samsung seems to follow the cautious policy of not introducing any dramatic changes into its new products. We can see improvements over the previous models (for example, over the 250GB M5S), but server applications are clearly not for this HDD.
Here are the same diagrams for the other tested drives:
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 with one thread, but it is the drive from Western Digital that takes first place with multiple threads. Moreover, this is the only drive that slows down less than twofold. The Toshiba suffers the biggest performance hit among all the drives when reading multiple threads. As for the newcomers, they are good enough here: the Hitachi is somewhat better overall but its speed plummets suddenly at four threads.
Samsung’s performance is higher than that of the Toshiba with one write thread although we might have expected the opposite after the sequential writing test. The Samsung even outperforms the Western Digital and gains the lead when processing multiple threads. The Hitachi copes with two threads well enough but has problems handling more threads. It shares last place with the Toshiba when writing three or four threads.
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.
As you may have expected, the HDD from Western Digital takes first place. There is tough competition among the other drives. For example, the Toshiba is second at low loads while the Samsung pushes the Hitachi down to last place. Then, the Hitachi overtakes the Toshiba at high loads and the Samsung sinks down to last place due to the low efficiency of its firmware algorithms.
Let’s check out the performance ratings.
So, second place goes to the Toshiba while the Hitachi and Samsung have nearly identical results.
Western Digital’s first place is unchallenged in this pattern, too. As for the other drives, the Hitachi is the slowest under low loads, but outperforms its opponents at high loads, trying to catch up with the leader. The Toshiba seems to have flaws in its firmware. Its graph does not go up. Instead, it has a noticeable slump at medium loads. As a result, this drive is slower than the Samsung under most loads, but equals and even beats the latter at very short and very long queue depths, respectively.
So, the Hitachi enjoys second place in the performance ratings while the Samsung is third, pushing the Toshiba back.
The Hitachi drive does not like workstations: it is a clear loser in this test. Interestingly, the Toshiba is almost as fast as the leader under very low loads, but loses its ground and gives way to the Samsung at long queue depths.
The performance ratings are explicable: the Toshiba is better than the Samsung because the results at short queue depths have bigger weights in our formula.
The standings are different when the test zone is limited to 32 gigabytes. The ex-leader Western Digital finds itself in last place. The Samsung and Toshiba are contending for first place at very short queue depths but the Toshiba gives up and is replaced by the Hitachi as the load grows high.
So, three drives have similar performance ratings, the Western Digital being noticeably slower than the leaders. The Samsung wins this test by a narrow margin.
We will use WinBench 99 to record the drives’ data-transfer graphs:
The next diagram compares the read speeds of the drives at the beginning and end of the partitions created on them.
Samsung’s HDD delivers the highest performance. Its platters seem to be just as dense as the Hitachi’s, and nothing should prevent the company from increasing the platter capacity from 160 to 167 gigabytes. The Toshiba shows the lowest speed both at the beginning and end.
Now let’s see what we have in the WinBench 99 tests performed for a 32GB partition. The results are sorted by the High-End Disk WinMark score.
WinBench 99 FAT 32 results
This rather old benchmark prefers the drive from Western Digital in the High-End part. But in the Business part the top position is shared by the Hitachi and Samsung.
WinBench 99 NTFS results
Neither Hitachi nor Samsung is brilliant in NTFS but the Toshiba drive goes neck and neck with the Western Digital in the Business subtest.
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.
We will only discuss the NTFS data because the standings are generally the same in FAT32. You can use the link below to view the full results: FC-Test FAT32.
As you may have guessed after the previous tests, the HDD from Western Digital is the best at creating files. The Samsung tries to compete with the leader on very large files only. On the other hand, the Samsung has improved the impression created by its poor performance in the sequential read test. FC-Test is closer to real-life loads than IOMeter, after all. Interestingly, there are no obvious losers: the drives are comparable to each other overall.
The HDDs from Western Digital and Samsung go on fighting with each other at reading. The former is somewhat faster with small files while the latter is ahead in the large-file ISO pattern – the Western Digital is only third then.
The Samsung takes first place when copying within the same partition while the Western Digital is competing with the Hitachi to be second. The Toshiba is obviously slower than the others.
When copying from one partition to another, the Samsung is challenged by the Hitachi. The Western Digital is very close to the leaders whereas the Toshiba is the slowest drive with nearly every file-set.
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.
Well, the Hitachi 5K500 was far from brilliant in the synthetic benchmarks, but seems to prefer real-life loads. At least it is the fastest drive when booting the OS up.
The Hitachi is also good at loading applications. This time it is closely followed by the Western Digital, though.
The Hitachi is in the lead when copying files, too. It is a favorite of this benchmark. The HDD from Western Digital is close to the leader while the Toshiba is slow again.
The Hitachi is first and the Western Digital is second again.
According to the overall scores, the Hitachi is first with a small lead over the Western Digital. The Samsung takes third place, being barely ahead of the Toshiba.
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 Hitachi is in the lead again. The Samsung is now closer to the drive from Western Digital but cannot overtake it.
We’ve got the same results in the Application Loading subtest: the drives from Western Digital and Hitachi are competing for first place and the latter wins again.
There is nothing new again: the standings are the same as in the previous version of this benchmark.
We’ve got the same leader but the other standings are different. The Samsung takes second place, being just a little faster than the Toshiba. The HDD from Western Digital has fallen far behind the others. Such sudden changes in the ranks are due to the fact that the drive’s caching algorithms affect its performance in this subtest. Most of the test is performed within the buffer as is indicated by the high data-transfer rates.
You might have predicted these results: Western Digital’s HDD has been showing excellent writing capabilities throughout this test session. The Samsung has modest performance here. It has very specific deferred writing algorithms and its write speed depends greatly on the type of load.
The HDDs from Hitachi and Western Digital share first place in this version of PCMark (it is good for WD that file writing is included into the benchmark as a subtest). The Samsung has the lowest overall rating.
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.
The HDDs from Western Digital and Hitachi are in the lead again.
The HDD from Western Digital wins the Gaming subtest. The Toshiba and Samsung are equals just like in the previous subtest.
The Hitachi copes best with loading image files. Loading something seems to be the job this drive likes the most. Take note that the Western Digital is in last place in this test, and the gap is quite large. So, there are some things this HDD cannot do well. We are not sneering. We just want to remind you that there are no perfect things in this imperfect world.
Booting up Windows Vista seems to be greatly different from booting up Windows XP. The ex-leader Hitachi has given way to the Western Digital here.
The Hitachi wins the Movie Maker subtest, though.
If you read our HDD related reviews often, you should know that this subtest is full of surprises as some drives manage to execute most of it within their cache buffer. This time it is the Samsung. Judging by the numbers, the drives from Western Digital and Hitachi processed some data of this test in their cache memory, too. Their results are not as high as that of the Samsung, though.
The Toshiba is surprisingly the best drive when loading files into Media Player.
This subtest seems to be revised greatly in this version of PCMark. The ex-leader Hitachi is only third now, and the Toshiba wins with a large advantage.
However, the Hitachi is the overall winner in PCMark Vantage, too. The Samsung and Western Digital share second place while the Toshiba has the lowest overall score.
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.
You might have supposed that Hitachi’s model would have the highest power requirements due to its three-platter design. However, it is the Western Digital that has the highest startup current. It probably needs such a high current to spin up its spindle and get ready to work faster.
Well, the time the HDD takes to spin up is unimportant for mobile storage devices whereas the maximum current is a factor that must be taken into account. As we know, the USB interface provides a current of 0.5A per connector. Many modern mainboards provide even more, but there is a limit anyway. Many users already had this problem when their USB drive just would not start up because of the lack of current for that. This problem was solved by using a power cord that could be plugged into two USB ports but this only gives you 1 ampere while Western Digital has stepped beyond that point. Will we see USB power cords with three connectors now?
The Toshiba has low power consumption in idle mode, and the other drives have similar results. Take note that the Hitachi needs a little bit more power than the others: its motor has to rotate three instead of two platters!
The drives needs about the same amount of power to do random reading. Well, this should have been expected since their platters are rotating, their heads are moving about, but their electronics is idle – all under similar conditions. The drives from Hitachi and Western Digital save less power at writing than their opponents as they perform deferred writing. The more advanced electronics of those HDDs just needs somewhat more power to do its job.
Here, not only deferred writing but also look-ahead reading is at work. Therefore the drives have about the same power draw in both modes. It is clear once again that the excellent speed of the Western Digital is accompanied with high power consumption. This extra half of a watt may be a problem for a mobile storage device. To remind you, the maximum current of 0.5A across a 5V bus equals 2.5 watts, which is lower than the 3 watts required by this HDD. And you should also take the consumption of the USB-SATA bridge into account.
We’d like to write the first paragraph about the Hitachi Travelstar 5K500. Its three platters with the now-highest recording density at a spindle rotation speed of 5400rpm is something to be counted with. The large dimensions of the HDD are a downside, but there are few alternatives to it today. There is one 500GB drive from Samsung but we have not yet seen it alive. The Travelstar 5K500 will be good in a server if the load is not mixed but close to pure reading or writing. It will also make a rather fast and very voluminous disk subsystem for a notebook (half a terabyte!) or can be installed into an external rack to provide uncompromising storage capacity. You just have to make sure beforehand that it will fit into your notebook or rack.
The 320GB model from Western Digital is the fastest 5400rpm drive today. It can challenge 7200rpm models, being only inferior to them in access time. Western Digital wrote firmware with excellent algorithms that make good use of the 16-megabyte cache buffer. The only downside of this drive is that it consumes considerably more power than same-class products.
The other two HDDs need to have their firmware revised. The Toshiba MKxx52GSX also seems to have slow mechanics as is indicated by its low performance in most of our synthetic tests. The Samsung SpinPoint M6S left us perplexed: it showed the best speed of reading, good results in FC-Test, excellent multithreaded writing, but also very poor performance in server applications and in PCMark. It seems to have high untapped potential. Hopefully, Samsung’s 500GB model will be much better and effect a small breakthrough on the market of compact HDDs.