|
Advice Beginners BIOS Guide CPUs Links Mainboards Memory Network Storage Video/Sound Cards Contact Forum SiteMap Sponsors WebNews Home |
. | . |
Prices: Mainboards ABIT ASUS Chaintech Shuttle Soyo Tyan CPU Intel P4 2.4C-800 P4 2.6C-800 P4 2.8C-800 P4 3.0-800 P4 3.2-800 AMD AthlonXP XP 1700+ XP 2000+ XP 2400+ XP 2500+ XP 2700+ XP 3000+ XP 3200+ Athlon64 Athlon64 3200+ Athlon64 FX-51 Opteron Opteron 240 Opteron 242 Opteron 244 Opteron 246 Memory Corsair Crucial Kingston Mushkin OCZ |
LOSTCIRCUITS |
||
| Fujitsu MHT2060AH SA, HighPoint RocketRAID 1640 and Level5 RAID All in an XPC | ||
|
(Review by MS, April 23, 2004) | ||
|
Fujitsu MHT2060AH ATA-100 |
Single Drive Performance
In this case, we are not going to go into gory detail, the purpose of this setup was not to look into single drives but what can be done with some micro-RAID configurations. Nonetheless, we wanted to give some basic drive parameters to put things into context and for this purpose, HDTach 2.61 is just fine except for the limitation of the read burst display. Unfortunately, the patch did not work for one reason or the other, however, the real value of the burst transfer is also displayed at the end of the HDTach logfile. For comparison reasons, we also ran HDTach 2.70, which uses far fewer sampling points and is overall less accurate and apparently even more prone to speedmatching conditions and lack of detection of errors but the general message is going to be the same for both versions. .
Basic Performance on ICH5
(the HDTach tests were run with the OS installed on the Seagate Barracuda 7200.7 and the Fujitsu drive freshly formatted and without data)
We first tested the Fujitsu MHT2060AH SA on the ICH5-R SATA controller of the ASUS P4C800E to take a look at the effective media transfer rate as well as the effective host transfer rate. For a 2.5" drive spinning at 5400 rpm, the sequential performance is no less than stunning. It gets even better when we are looking at the burst reads, with 114 MB/sec, the MHT2060AH SA does not need to shy any comparison with any "grown-up" 3.5" HDD.
Within the inaccuracies of the benchmark utility, the results with ver 2.61 are identical to those shown above, the burst transfer was logged as 112 MB/sec in this case
HighPoint RocketRaid 1650 single drive performance
The sequential performance is the same as that we saw on the ICH5-R, which is natural since the platter geometry and particle density are the only factors that define effective media performance. With approximately 32 MB/sec sequential read speed at the drive's OD, the MHT2060AH SA shows that there is no reason to believe that a 2.5" drive has to be slow. What is interesting, though is the fact that the burst transfer is stunted at about 65 MB/sec. The same data were obtained on both test systems. In theory, we would have expected at least some burst transfer rates in the order of 80-90 MB/sec, which is still way below the limitations of the PCI bus.
Promise FastTrak S150 TX4 single drive performance
To check for system anomalies, we ran the same drive on the Promise FastTrak S150 TX4 with the expected result that the sequential performance is once again identical to that found with the two other controllers. Interestingly, though, the burst rate is with roughly 85 MB/sec where we would have expected it to be for a PCI-based bridge controller
With respect to the burst transfer numbers shown above, keep in mind that these numbers reflect the measuring criteria used by HDTach. Those numbers are not necessarily indicative of what is going on in real live as we'll show later. In fact, we have been getting transfers in the order of close to 100 MB/sec with the HighPoint controller, but under the conditions used by HDTach, the highest we were able to obtain was 65 MB /sec in a single drive configuration.
next page: => RAID Level0 performance =>
If you enjoyed reading this article and found it useful, please consider making a small donation to LostCircuits.