by Aleksey Meyev
02/24/2009 | 11:20 AM
Moor’s law states that the amount of transistors in a single electronic chip doubles every two years. This concept is often applied to other computer products whose characteristics increase at the same rate. For hard disk drives, it is storage capacity. 2.5-inch HDDs have even been developing ahead of the schedule recently. About two years ago 160GB models were introduced and then 250GB models came out half a year later. And now we are ready to test 500GB products based on two platters. Thus, the recording density has grown from 80GB to 250GB per platter in two years.
This progress was made possible by perpendicular recording technology. Replacing longitudinal recording, it has helped HDDs grow up in recording density and capacity. In fact, it is for 2.5-inch HDDs that the increased recording density is the most beneficial because such products cannot utilize a large number of platters to offer more storage space.
Currently, there is a race among the manufacturer for maximum-density platters because high-capacity HDDs help attract the customer and claim a leadership in the industry. By the way, the tight competition has made 2.5-inchers closer to 3.5-inch models in terms of pricing. Right now, 500GB 2.5-inch drives are less than three times as costly as cheap 3.5-inchers of the same capacity. The performance of today’s 2.5-inchers is high enough thanks to high recording density. Their capacity is sufficient for most applications and they are quieter than 3.5-inchers. So, today we are going to discuss 2.5-inch 500GB hard disk drives with a spindle rotation speed of 5400rpm.
The first model of that capacity to come to our labs was Hitachi Travelstar 5K500. Alas, it is based on 167GB platters and thus has three rather than two platters in total. As a result, it is thicker. 3-platter HDDs have a thickness of 12.5 mm while 2-platter ones are 9.5 thick. Not all notebooks and external HDD enclosures can accommodate a thick drive, so the Hitachi 5K500 has limited applications in comparison with thin models. When it comes to speed, this model is about as fast as 320GB ones, as we learned in one of our previous articles. Anyway, it must be given the credit of being the first ever small 500GB HDD.
This HDD perplexed us a lot. It was announced at the manufacturer’s website at the same time with the Hitachi 5K500 but its thickness was 9.5mm, which meant that it had two platters. And it is these two platters that we could not believe in then. Samsung did not declare the number of platters in the product specs (and does not declare it now – the platters/heads statistics in the specs table below is given according to our own examination). We should confess we could not decide at first whether this HDD had new platters or the manufacturer somehow fitted three platters into a thin case.
The 500GB model fitted nicely into Western Digital’s Scorpio Blue series. Interestingly, every other model from this series has a Scorpio Black counterpart that features a spindle rotation speed of 7200rpm and a larger cache.
As an opponent from the previous generation, i.e. with a storage capacity of 320GB, we will take a Western Digital Scorpio Blue which came up the best in our comparative test of six 320GB 2.5-inchers.
The following table lists the specs and firmware versions of the HDDs.
The following testing utilities were used:
Testbed configuration:
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. If necessary, the HDDs were switched from quiet (with Advanced Acoustic Management enabled) into ordinary operation mode.
We use our internal tool called IOMark for low-level tests of hard disk drives.
The following diagram compares the HDDs’ read speeds at the beginning and end of the partitions created on them.
Yes, this is the test that proves that the Samsung drive has a higher recording density than the HDDs with 160GB platters. The advantage is especially conspicuous at the end of the partition. The 500GB model from Western Digital is even better, though. Its read speed is higher by almost a third – just as its recording density is higher by a third!
Next let’s see the HDDs process data in their cache. You can click the following links to view the graphs while we will discuss a diagram showing the top speeds:
Every HDD reads from the buffer somewhat faster than writes to it although the difference is very small. No HDD has problems with large data chunks. The 500GB model from Western Digital is in the lead, notching a burst read speed of over 200MBps. It is followed by the Hitachi that has a burst speed of almost 170MBps. The Samsung and the 320GB Western Digital are only as fast as 100MBps. This must be due to the fact that the HDDs came with SATA150 mode enabled.
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 and we can see the dependence of a drive’s sequential read/write speed on the size of the processed data block. This test is indicative of the maximum speed a hard disk drive can achieve.
The numeric data can be viewed in tables (see the links below). We will be discussing graphs and diagrams.
IOMeter: Sequential Read results
This test proves again that the Samsung has got higher-density platters. The 3-platter Hitachi is almost as fast as the 320GB from Western Digital while the Samsung is far faster on large data blocks. However, it has the lowest speed among the tested HDDs on small data chunks. Besides, the Samsung accelerates to its top speed on 64KB data blocks whereas the others deliver their top speeds starting from 8KB blocks.
The 500GB drive from Western Digital is very good. It is very effective at gluing together small-size requests and delivers a huge top speed (for a 2.5-incher). Just a couple of years ago the read speed of over 80MBps could only be achieved with the then-best 3.5-inchers.
IOMeter: Sequential Write results
The Samsung is poor at writing. Its results are the same as those of the lower-density models. The 500GB Western Digital is again good at processing small data chunks and has a high top speed.
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 read response time depends not only on the recording density and rotation speed of an HDD’s platters, but also on the speed of the heads, the number of servo marks per platter and some other important factors. As a result, the Hitachi has a very slow response. The Samsung is not very good, either, and is 0.5 milliseconds inferior to the 320GB Western Digital. The new 500GB model from Western Digital is superb in this test: its read response time of 16 milliseconds is very, very good.
The average write response depends mostly on the efficiency of deferred write algorithms. And again, both drives from Western Digital are in the lead: excellent mechanics is backed up by good firmware algorithms in the 500GB model which wins this test. The opponents are worse but not hopeless.
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 in two ways. 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 where the traditional measurement of speed in megabytes per second becomes more relevant.
Let’s start with reading.
IOMeter: Random Read, operations per second
The speeds are directly proportional to the HDDs’ access time, so the 500GB model from Western Digital is ahead of its opponents.
IOMeter: Random Read, megabytes per second
When the data chunks are larger, the sequential read speed becomes the decisive factor. This helps the Samsung leave the models with 160GB platters behind and take second place.
IOMeter: Random Write, operations per second
Western Digital’s products have no rivals when writing in small data blocks. As for the losers, the Samsung is slower on the smallest data chunks but the Hitachi slows down on larger ones.
IOMeter: Random Write, megabytes per second
The HDDs from Western Digital are superior when writing in large data blocks, too. This manufacturer’s 2.5-inch HDDs have been very fast at writing for a long time. People at Western Digital seem to know some secret.
The Samsung and Hitachi are competing still: the Hitachi is on the losing side through most of the tests but the higher-density Samsung suddenly loses its ground on very large data chunks.
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 for IOMeter: Database pattern.
We will build diagrams for request queue depths of 1, 16 and 256.
When the queue is only 1 request deep, an HDD has nothing to reorder. It can only process read requests as fast as it can and put write requests into its cache. The 500GB model from Western Digital is superior at both and enjoys a nice lead over its 320GB cousin. The Samsung has less effective deferred writing algorithms: its performance does not grow up much as the percentage of writes increases. The Hitachi is different: it obviously has very effective deferred writing algorithms but it does not like mixed loads (with about the same share of reads and writes).
NCQ algorithms come into play at a queue depth of 16 requests. Western Digital’s drives feel all right and both behave in a similar way. Judging by the performance growth in the left part of the diagram, these HDDs have NCQ. The Samsung still shows poor deferred writing and has no request reordering. The Hitachi is more effective than the Samsung at pure reading and writing but is very poor under mixed loads. So, it is hard to tell which one is the worse of the two.
There is nothing exciting when the queue depth is increased further. Western Digital is still in the lead. Interestingly, the lower-density 320GB model is better at high percentages of reads – it seems to have more effective NCQ algorithms. The 500GB model is better at processing write requests, though.
The other two HDDs are not quite good: the Samsung prefers to process requests as they are, without any optimizations, and only applies some deferred writing to improve performance. The Hitachi shows excellent deferred writing and good NCQ algorithms but slows down under mixed loads.
The following diagrams show the performance of the HDDs with 250GB platters at five request queue depths.
Indeed, the Samsung has no NCQ at all. Its performance is nothing but mediocre in this test. Its 320GB cousin was no record-breaker, either, but was faster in server tests and had some kind of NCQ.
Western Digital’s firmware for 2.5-inch drives is nearly ideal. The 500GB model has inherited excellent NCQ algorithms and has even improved somewhat in terms of deferred writing.
You can find such diagrams for the other two HDDs in the above-mentioned reviews.
The HDDs will be 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.
Western Digital’s products are beyond competition at short queue depths but then the Hitachi gets close to them thanks to its excellent NCQ algorithms. Interestingly, the firmware of the 500GB Western Digital has got worse in some aspects: the 320GB model goes ahead at long queue depths.
The Samsung is obviously no good for servers with random-read load.
The two HDDs from Western Digital are contesting for top place again. The newer model wins at short queue depths while the 320GB model, at long ones. The Samsung is rather good at short queue depths because this pattern contains write requests, but does not speed up as the queue is getting longer. As a result, it is eventually outperformed by the Hitachi, which was very poor at short queue depths.
Our formula gives higher weights to short queue depths, therefore the 500GB Western Digital has the highest score while the Samsung and Hitachi are almost equals.
The 500GB Western Digital is the best drive for workstations, and the 320GB model actually has no chance here. The Samsung feels good thanks to the large portion of writes in this pattern and nearly catches up with the 320GB Western Digital. The Hitachi fails the test.
We’ve got the same leader when the test zone is limited to a 32GB partition. The 3-platter Hitachi is surprisingly second here.
The performance ratings show the same standings as in the diagrams.
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 and the number of outstanding requests from each client varies from 1 to 8. We will 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:
Well, each drive is quite sensitive to the number of read threads. Western Digital’s products are better than others but the 500GB model slows down suddenly when processing four threads. The Hitachi behaves in the same manner but its performance hit at four threads is terrible – it has only one tenth of its single-thread speed then. The Samsung, on the contrary, slows down at two threads but then does not change much at three or four threads.
The HDDs have the same amount of cache memory and should not differ much in terms of caching and writing. However, Western Digital’s products are good (note that there is about the same gap between them at every load). The Samsung copes very well, being ahead of the 500GB Western Digital with multiple threads. Having showed an excellent deferred writing in previous tests, the Hitachi is unexpectedly the worst drive when writing multiple threads simultaneously.
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 HDD’s behavior under complex load. In fact, the HDD is working with two threads (one for reading and one for writing) when copying files.
We will discuss the NTFS results which are overall the same as the FAT32 ones as you can make sure by clicking the following link.
Western Digital’s HDDs are very good at writing as is reconfirmed by FC-Test. Take note that the 500GB Samsung, with higher-density platters than the 320GB Western Digital, can only beat the latter on very large ISO files. The 500GB Western Digital copes well with any files and is unrivalled in this test. The Hitachi is hopelessly slow in comparison with the leader.
The new drive from Western Digital wins the test of reading, too. The Samsung and the 320GB Western Digital are competing for second place: the Samsung is better with large files whereas the 320GB Western Digital is ahead with smaller ones. The Hitachi is the slowest drive again, but the gap is very small now.
The 500GB model from Western Digital wins yet another test. Then the standings are rather unusual due to the peculiarities of the HDDs’ firmware. The 320GB WD is now competing with the Hitachi, which has not been fast in the previous subtests, whereas the higher-density Samsung takes last place. Its firmware must be poorly optimized for such load.
The Samsung jumps from last to second place when the copying is done between two partitions. The HDDs are similarly slow with small files, though. The difference can only be seen with large files.
PCMark04 benchmarks hard disk 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’s 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.
Each test is performed ten times and the results are averaged.
The new Western Digital is better than others at booting the OS up. This HDD seems to be going to win every test possible! The Hitachi is good, taking second place. The Samsung turns in a disappointing performance.
The Hitachi is the best at loading applications. This manufacturer seems to know some secret that helps beat the higher-density 500GB Western Digital. The Samsung shows a poor performance again.
The 500GB Western Digital wins the test, enjoying a 20% over its opponents.
The Hitachi steps up on top again in the General Usage test. Interestingly, Western Digital’s 320GB model is ahead of the 500GB one. There must be some differences in the drives’ firmware indeed.
So, first place goes to the 500GB Western Digital but the Hitachi is very close to it. The Samsung is a disappointment, being much slower than the best models of the previous generation.
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.
This version of PCMark shows that the Hitachi can boot Windows XP faster than the other HDDs. The Samsung is again slow.
The standings are the same as in the previous version of the benchmark: the Hitachi is ahead, the Samsung is far behind the others.
The same goes for the General Usage test: the Hitachi wins while the new 500GB drives are slower than the leaders.
The 500GB Western Digital enjoys a huge lead in the Virus Scan test. Its 320GB cousin finds itself in last place: their firmware differ indeed. The better firmware algorithms of the Hitachi help it outperform the higher-density Samsung.
There are no surprises at writing: Western Digital’s HDDs meet no competition. There is not much work for firmware algorithms here and the Samsung is better than the Hitachi thanks to higher recording density. However, the Samsung is inferior to the lower-density 320GB Western Digital. So, recording density is not the only performance-determining factor even under such simple loads.
The 500GB Western Digital is now an absolute winner. The Samsung is poor again: its dense platters are hamstringed by ineffective firmware.
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.
Western Digital’s products are faster than their opponents under multithreaded load in which one thread is scanning files. And the newer 500GB model is faster than the older 320GB one. The Samsung with inefficient firmware takes last place again.
Western Digital’s HDDs are the best choice for a gaming computer.
However, they are very poor when it comes to processing photographs. The Hitachi wins this test with a large lead.
The Blue Scorpios return to top places when booting Windows Vista up.
The 500GB Western Digital is better than others for video-editing applications. The Samsung is a disappointment again: its high recording density can’t do anything without effective firmware algorithms.
For the Media Center test it is recording density that is the decisive factor. As a result, the two 2-platter models take top places with similar results.
The HDDs produce similar results in the Media Player test.
This time, the Samsung is closer to the others in the Application Loading test. The 500GB Western Digital is the winner now.
So, the 500GB Western Digital wins the test but the 500GB Hitachi with its three 167GB platters is quite good under typical home PC applications. The Samsung is far worse than both leaders.
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 take a look at each mode individually:
We used to criticize Western Digital for the aggressive spinning-up of the platters in its products. Yes, the company’s HDDs would spin up faster than opponents, but required such a high current that many users had problems powering them via USB. We can remind you that the maximum current through a USB connector is 0.5A (although specific mainboards can provide more). Thus, a HDD can get only 1 ampere from two USB ports.
It is good that Western Digital has thought about that problem: while the 320GB model consumes more than 1 ampere at startup, the 500GB model is the most economical in this review. The Samsung is economical, too. The 3-platter Hitachi needs a little more than 1 ampere, so you can have problems powering it from USB ports.
When not accessed, three of the four HDDs deliver similar results and require less than 1W of power. The Samsung is somewhat worse than the others here.
Take note that today’s drives do not spend much power for keeping their platters running. The Hitachi needs about the same amount of power for its three platters as the others for two.
Random writing requires less power in comparison with reading. The higher consumption of the electronics at writing seems to be made up for by the deferred writing algorithms in each drive.
The new 500GB Western Digital is overall somewhat more economical than its 320GB predecessor. The difference is 0.28W at both writing and reading. This does not look like a lot, yet amounts to about 10%. It is especially good that the new Scorpio is the only drive to require less than 2.5W (the maximum amount of power provided by a single USB port).
As for the other participating devices, the 2-platter Samsung needs somewhat more power than the 3-platter Hitachi. Easy to guess, this depends on the power consumption of the drive’s actuator.
It is at sequential operations that a HDD usually has its highest power consumption because both the heads and the electronics are working at their full capacity. This test is an example of that. Take note that the HDDs generally need more power at writing and at reading. This is due to the electronics’ caching requests while there is no opportunity for request optimization.
And again the 500GB Western Digital deserves our praise. While the previous model was very hungry, the newer model almost fits within 2.5W, catching up with the very economical Hitachi. The Samsung is voracious. Its electronics seems to be not optimized in terms of power efficiency.
By the way, all the power the hard disk drive consumes at work transforms into heat in its electronics or mechanics. Thus, the power consumption data are indicative of how much power the HDDs dissipate. The most economical HDDs are also going to be the coolest.
2.5-inch hard disk drives have grown by a third in terms of recording density: from 160 to 250GB per platter. Their storage capacity has increased proportionally, but what about their performance? This test session has given us the answer. It is the speed of sequential operations that has increased the most. Western Digital’s products (they make a good example because they have similar firmware) show an increase from 65 to 80MBps. This amounts to 25%, which is quite a lot. The only problem is that you can only see such performance benefits when reading or writing large files. At ordinary operations the drive’s firmware has such a great effect on its performance that the 160GB-platter models are often ahead of the Samsung, which has two higher-density platters but less effective firmware algorithms.
Talking about the specific products, the 500GB Scorpio Blue drive from Western Digital has inherited the good firmware of the previous models of the series. Besides, it has new electronics and has become more economical. As a result, this is a universal product that stands out among its opponents.
The Hitachi Travelstar 5K500 was not brilliant under server loads and at sequential operations (which is no wonder as it has three 167GB platters rather than two 250GB ones) but it feels all right under real-life loads. Its 3-platter design and increased thickness is its main downside. Hopefully, the company will switch to new platters soon and produce a worthy 2-platter opponent to the Western Digital.
The Samsung SpinPoint HM500LI is a disappointment. The huge potential of its platters is spoiled by ineffective firmware. Its electronics should be revised, too, in order to catch up with the opponents in terms of power efficiency. Anyway, this HDD is going to be popular because there are few 500GB 2.5-inchers available as yet. Its high storage capacity is a strong factor in favor of this product.
