DiskGraf Disk, LAN and USB Benchmark Results On PCs
See
Example Log File and
Results and Examples
Wide Page - This page was set up as 800/1000 pixels wide to accommodate preformatted text <PRE> results tables. Some browsers might produce
an unsuitable monospaced font for this, resulting in an even wider page. In certain cases, font size can be changed via browser Preferences.
Description
DiskGraf Benchmark measures disk writing and reading speeds at different block sizes, producing a graph of megabytes per second and CPU utilisation. It also does the same for bus/DMA speeds and measures random reading time of various sized files. Results are saved in a text file. This document is a summary of maximum speeds shown for a range of PC systems. Pre-compiled versions of the benchmarks can be found in DiskGraf.zip which also contains the source code and more detailed explanations.
The benchmark can also be used for testing disks on a Local Area Network and for CD/DVD drives with RW discs (see Results). See also CDDVDSpd Results.htm for disk and other device speeds at different file sizes and My Main Page for other PC benchmarks and results.
The benchmark writes a number of files (default 5, maximum 16) then reads them, each at block sizes 1 KB to 1024 KB. Default file size is 8 MB, but this can be increased to up to 10 GB. For compatibility with Windows 9X, CPU utilisation is measured via a second thread that soaks up free CPU cycles. This leads to the utilisation results only being valid on single CPU systems.
DMA speed is measured by repetitively reading the same block. Normally, this will be from the disk’s buffer, directly over the bus, to the CPU. These tests again use block sizes 1 KB to 1024 KB. Note that measured disk transfer rates will be less than those on the DMA test.
The random test measures random reading time in milliseconds of 1KB blocks out of 1, 2, 4 and 8 MB or larger (up to 1 GB, depending on maximum file size used for the other tests). As shown in the results, this helps in identifying the disk’s buffer size and RPM. On the latest disks, with buffer sizes of 8 MB or greater, the default file size has to be increased (menu option) to produce appropriate measurements.
The benchmark program uses FILE_FLAG_NO_BUFFERING in the CreateFile function to tell Windows not to keep the data in main memory based File Cache. Without this setting, data would generally be read from memory instead of from the disk. The program has an option to enable caching and show memory to memory data transfer rates.
A 64 bit version is now available in More64Bit.zip with minor changes described in 64 Bit Disk Tests.htm. These include a different method of measuring CPU utilisation, applicable also for dual CPU systems. Results are included below (see q64AMax1 - XP x64, jC2DSea2 - 64 Bit Vista, also yDiskUSB2) and, as one would expect, are usually similar to 32 bit version speeds.
Measurements include results for one of the fastest 2009 disks in an eSATA enclosure (system jC2DSea3). For further details
See Report.
Results for year 2009 include a WD low power Green Disk (qPheWD1 using 64-Bit Windows 7). WD do not publish the usual performance specifications such as RPM and latency for these disks . Instead, they just indicate IntelliPower, with repudiated marketing suggestions that the disk speed varies between 5400 and 7200 RPM. The consensus is that the speed is (was?) fixed at 5400 RPM. This benchmark indicates maximum writing speed of 94 MB/second and 104 MB/second reading, better than some recent 7200 RPM disks.
Extended Random Access Times suggest that the RPM might be somewhat higher than 5400.
Further 2009 disk results are for a Solid State Disk (SSD), where maximum reading speed of more than 200 MB/second could be limited by SATA 2 bus speed. Although this is much faster than current hard disk drives, writing speed is not as good, but constant random access time is superior at 0.1 milliseconds.
Example Results Log File
AMD XP 2.34 GHz, Hitachi 7K250 123 GB Disk, ATA100, 7200 RPM, ATA133 mainboard, Win2K
###########################################################
Disk Test DiskGraf Version 1.3A Tue Apr 20 20:59:56 2004
Copyright Roy Longbottom 2001
Windows NT Version 5.0, build 2195, Service Pack 2
CPU AuthenticAMD, Features Code 0383FBFF, Model Code 00000681, 2338 MHz
Memory 523760 KB, Free 419596 KB
F:Disk/partition 2051 MB, Free 2039 MB
Files 5 Size 8 MB Block Sizes 11 Minimum RAM Data
Notes: nVidia driver 4.12
Block File ->
KB 1 2 3 4 5
Write MB/Second
1 5.9 6.1 6.2 6.2 6.2
2 12.1 12.1 12.1 12.1 12.0
4 22.0 22.1 22.2 22.3 22.2
8 38.9 39.5 39.6 40.0 39.7
16 61.9 52.5 58.2 61.8 58.4
32 59.4 64.0 52.3 58.1 58.7
64 62.4 64.0 62.6 62.6 67.0
128 58.3 64.0 59.4 52.6 63.0
256 54.0 65.8 39.0 55.2 64.0
512 61.6 56.2 66.3 58.0 66.5
1024 59.1 50.1 62.1 58.2 52.1
Read MB/Second
1 10.7 10.8 11.0 10.7 11.0
2 19.6 20.4 20.0 20.1 20.4
4 29.8 30.1 29.4 29.6 29.5
8 47.5 48.9 49.6 49.0 49.1
16 51.0 52.0 52.1 51.6 55.2
32 53.6 52.2 52.3 54.6 55.2
64 52.4 52.3 52.3 52.3 52.4
128 51.2 52.7 52.6 55.7 52.7
256 51.4 53.4 53.3 53.3 53.4
512 54.2 55.3 57.8 55.0 55.1
1024 52.0 55.0 57.8 55.0 55.1
Write CPU Utilisation
1 18.6 16.9 17.1 16.7 16.9
2 16.9 16.1 16.2 16.4 16.6
4 15.4 14.8 15.5 14.9 15.3
8 15.8 13.5 14.4 13.9 13.6
16 10.8 8.9 9.9 10.6 12.0
32 7.6 7.5 6.3 8.7 7.0
64 2.5 2.5 2.4 2.8 2.6
128 2.1 2.2 2.2 1.6 4.4
256 2.1 2.1 2.0 1.6 2.0
512 2.0 3.6 2.1 1.5 3.9
1024 3.1 1.3 1.9 3.8 1.4
Read CPU Utilisation
1 26.5 26.8 27.9 26.7 27.5
2 24.4 24.8 25.1 26.2 24.9
4 17.9 18.1 17.7 18.7 18.2
8 15.4 14.9 15.1 16.8 15.1
16 7.3 7.5 9.9 9.5 10.2
32 5.6 6.4 5.5 3.7 3.6
64 2.6 1.2 1.2 2.5 1.3
128 1.1 1.5 2.4 1.3 1.1
256 3.3 1.0 1.0 1.1 1.1
512 1.2 1.2 1.3 1.6 1.2
1024 1.1 1.1 1.3 1.1 2.8
DMA Reread block up to 100 times at 1 to 1024 Kbytes
Read MB/Second
10.8 19.1 46.3 64.9 80.3 86.9 90.8 91.0 91.2 91.1 91.1
Read CPU Utilisation
27.0 52.5 30.2 40.8 12.4 6.5 3.0 2.8 3.3 3.1 2.9
Random Read 1KB out of 1, 2, 4, to file size Mbytes
Read Milliseconds
0.511 1.88 3.94 4.39 0.000 0.000 0.000 0.000 0.000 0.000 0.000
Read CPU Utilisation
4.0 2.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
End Disk Test Tue Apr 20 21:00:57 2004
|
To Start
Examples and Results
Results tables cover a range of disk drives providing data transfer rates between less than 1 MByte Per Second to greater than 200 MBytes Per Second, the highest being for RAID configurations. Speed is generally dependent on disk rotation speed (RPM) and amount of data on a track. However, it can be limited by DMA/bus speed with other variations due to bus driver used, mainboard, BIOS settings, Operating System version, disk space used, data fragmentation and, to a small extent, CPU and memory speed (these affect CPU utilisation). Comparison with the best results can help to identify problems, some of which are shown below.
CPU utilisation reduces with increased block sizes to close to zero percent on the latest processors, where measurement is not very accurate. The tables show CPU utilisation for block sizes up to 64 KB.
The original version had FILE_FLAG_WRITE_THROUGH in the CreateFile function. This made no difference in performance using Windows 9X but produced slower writing speed via Windows NT/2K/XP, with more significant differences on the later Operating Systems. The benchmark program was revised to remove this flag, the original being indicated by # in the writing speed table. Examples of differences are:
|
CPU Disk Writing MBytes/Second at KB Block Size
Key MHz RPM 1 2 4 8 16 32 64 128 256 512 1024
q64ASeaR Old 1991 15000 0.33 0.62 1.2 2.4 4.3 7.3 13 21 29 32 52
q64ASeaR New 1991 15000 7.4 13 20 39 60 82 106 108 112 133 133
q64AMax1 Old 2210 7200 7 10 19 21 23 26 27 31 31 33 34
q64AMax1 New 2210 7200 17 29 45 58 54 61 60 54 59 61 62
oThoHi1 Old 2080 7200 0.6 1.3 4.2 14 26 30 26 35 34 36 38
oThoHi1 New 2080 7200 6.5 12 22 36 45 52 47 46 47 47 46
iP4Sea2 Old 1900 5400 1.1 1.8 2.7 4.3 7.0 10 11 13 13 15 15
iP4Sea2 New 1900 5400 3.6 6.4 10 18 20 21 21 21 21 21 21
|
To Start
DMA Operation
Following are example of results that might be expected from the DMA test, which is measuring data transfer speeds from the disk’s buffer. Some examples show that speed can reduce at larger block sizes. It is not clear why this happens as the buffers are large enough to hold all the data and, sometimes, disk reading speeds can be faster than the lower figures. It must have something to do with repetitively reading the same data and, maybe, the disk reading ahead on the next track.
Two examples are given for q64AMax1. This has a 16 MB buffer and maximum data transfer rate of 66 MB/second. The latter, at 8.33 milliseconds rotation time, suggests a maximum track capacity of 550 KBytes. Yet it can not read the same 256 KB continuously.
|
CPU Disk MBytes/Second at KB Block Size
Key MHz RPM 1 2 4 8 16 32 64 128 256 512 1024
gp3SeaA 1400 7200 7 11 15 20 23 25 26 27 26 20 16 UDMA 33
nTbdIBM8 1400 7200 8 17 33 43 50 56 59 60 60 60 60 UDMA 66
nTbdQua 1200 7200 6 11 21 35 53 65 74 77 81 84 83 UDMA 100
oThoHi1 2080 7200 12 25 48 71 78 87 91 93 93 93 93 UDMA 133
q64AMax1 2210 7200 22 39 60 85 104 118 125 131 41 29 38 SATA
q64AMax1 2210 7200 23 38 60 85 104 118 126 131 34 53 59 SATA
jC2DSea1 2400 7200 15 18 35 68 109 145 173 208 229 237 243 SATA2
ip4WD2 2411 7200 15 27 40 57 70 79 84 87 88 30 40
q64ASea1 2150 7200 9 18 32 49 65 77 86 91 89 30 40
|
To Start Results/Examples Index
Random Reading
As indicated above, the random test reads 1 KB blocks out of 1, 2, 4 MB or higher, the upper limit depending on the file size chosen for other tests. When this fits within the disk’s buffer, the reading speed is reasonably fast. When the size is greater than the buffer (or some smaller default), the average reading time can be expected to be slightly higher than half the disk revolution time. Here, the milliseconds for the various disk RPM specifications are:
4200 - 7.1 ms 5400 - 5.6 ms 7200 - 4.2 ms 10000 - 3.0 ms 15000 - 2.0 ms
Following are examples from the later detailed table to show what can be expected. With the 8 MB buffer, a lot of accesses are from the disk, but a test run with 16 MB files is really required.
|
Buffer CPU Disk Milliseconds at MB file size
Key MB MHz RPM 1 2 4 8 16 32 64
nTbdMax1 2 1533 5400 2.0 4.4 6.3 6.8 7.5
mDurMax2 2 950 7200 0.8 2.6 4.3 4.9
q64ASea1 8 2150 7200 0.4 0.8 2.0 3.9
q64AMax1 16 2210 7200 0.04 0.04 0.2 0.9 2.9 4.1 4.9
|
To Start Results/Examples Index
File Cache Operation
The following results show example reading speeds when the benchmark option is chosen to allow Windows to cache the data in memory. Writing speeds tend to be slightly slower than normal, when using this option. The table shows that there can be variations on the same PC, depending on the version of Windows, and speed can reduce with larger block sizes, as on the DMA test.
|
CPU Disk MBytes/Second at KB Block Size
Key MHz RPM 1 2 4 8 16 32 64 128 256 512 1024
mDurMax1 W98 950 7200 132 168 137 175 198 210 153 140 141 140 144
mDurMax2 W2K 950 7200 216 244 260 266 262 250 159 156 156 156 156
iP4Sea2 1900 5400 340 455 554 633 682 706 703 388 263 266 262
iP4Max 2400 5400 293 427 406 632 794 907 1004 1004 467 390 391
q64AMax1 DCDR 2210 7200 690 1016 1292 1519 1630 1619 1601 1591 1352 840 838
jC2DSea2 DDR2 2400 7200 420 717 1089 1512 1816 2000 2233 2331 2358 1995 2064
|
To Start Results/Examples Index
PCI Bus and SCSI Operation
The standard 33 MHz PCI bus has a maximum speed of 133 MB/Sec and 93 MB/sec has been recorded with UDMA 133 on the DMA tests. As can be seen in the reading speed table, CPU utilisation using SCSI is much lower than for IDE disks. So, it comes as rather a surprise that PCI/SCSI disk reading speeds (and via the DMA test) have been at 60 MB/Sec or less. Except for a RAID0 set up, this did not matter much until recent times, when a single disk could achieve this data transfer speed.
The following show reading speeds on a 15K RPM disk (R0) and two in RAID configurations, all via an Ultra 160 controller. The tests were run at the different PCI bus speeds shown and results were similar to those measured on the DMA test. The disks have maximum rated speeds of 68 MB/Sec. Calculations indicate that the PCI/SCSI overheads for each block transferred are higher than the data transfer time. See later results for fast speeds via 133 MHz PCI-X.
|
Bus CPU Disk MBytes/Second at KB Block Size
Key MHz MHz RPM 1 2 4 8 16 32 64 128 256 512 1024
oPalSeaR0 50 1700 15000 2.5 5.3 10 19 29 44 44 56 57 57 57
oPalSeaR2 34 1700 15000 1.7 3.6 7.0 13 21 30 40 47 51 56 58
oPalSeaR2 42.5 1700 15000 2.0 4.4 8.4 15 26 38 48 59 64 68 70
oPalSeaRd 50 1700 15000 2.2 5.0 10 18 29 43 57 64 73 79 82
|
To Start Results/Examples Index
Speed Over Whole Disk
Modern disks do not have the same amount of data on each track, but the surface is divided into zones which do have equal track capacities. The inner zone might have half the amount of data than in the outer zone and transfer rate will at half speed. Following are two examples of measurements in different partitions, where the last results are at the 92% and 95% point.
|
CPU Disk MBytes/Second at KB Block Size
Key MHz RPM 1 2 4 8 16 32 64 128 256 512 1024
iP4Sea2 50% 1900 5400 6.0 11 15 21 28 28 28 28 28 27 28
iP4Sea2 42% 1900 5400 6.2 11 14 24 24 24 24 24 24 24 24
iP4Sea2 8% 1900 5400 6.2 11 19 19 19 19 19 19 19 19 19
oPalIBMRd 22% 1396 7200 11 18 26 29 46 65 68 63 61 58 62
oPalIBMRd 71% 1396 7200 12 20 30 39 55 56 63 63 62 58 62
oPalIBMRd 5% 1396 7200 11 19 35 34 36 35 34 35 35 33 31
|
|
Speed can be slower than expected if a disk or partition has lots of data. The data might also be fragmented and a benchmark file could be scattered over a wide area, producing slow results. This is usually indicated by variations in speeds on the five (or whatever) files used for testing. At one time, tests showed that Windows wrote files in the first unused space. You could nearly fill the disk then delete almost all of the files but the next files saved would be on the slower part of the disk. This might still apply.
On a modern fast disk, maximum speed might not be shown with 8 MB files. The following shows the difference between reading 8 MB and 32 MB files, where the maximum transfer rate is greater than 100 MB/sec on a RAID system.
|
CPU Disk MBytes/Second at KB Block Size
Key MHz RPM 1 2 4 8 16 32 64 128 256 512 1024
iP4WDSRd 8MB 3633 10000 14 23 37 53 66 53 88 83 100 104 102
iP4WDSRd 32MB 3633 10000 14 24 37 54 68 65 89 103 108 109 108
|
To Start Results/Examples Index
Write Caching
Depending at least on the bus driver and disk buffering strategy, sometimes writing speed can appear to be faster than reading and producing an apparent higher data transfer speed than the disk’s specification. This is due to Windows informing the benchmark program that transfer has been completed when data has been sent to the disk’s buffer.
The following shows measured writing and reading speeds on a SATA RAID0 system. This has two disk drives with 8 MB buffers and the maximum drive speed rating is around 60 MB/Sec. Default file size of 8 MB is used.
|
CPU Disk MBytes/Second at KB Block Size
Key MHz RPM 1 2 4 8 16 32 64 128 256 512 1024
iP4HiSRd Write 3000 7200 5.0 10 19 35 56 69 82 110 155 166 168
iP4HiSRd Read 3000 7200 6.4 18 29 47 69 55 67 68 85 105 101
|
To Start Results/Examples Index
USB Disk Drives
The following show writing, reading and DMA speeds of disk drives via high speed USB 2. Three of the examples are for the same Hitachi 7200 RPM disk drive, FAT32 formatted, which is probably capable of achieving 60 MB/Second, same as the HS USB 2 maximum speed. In all cases, reading speed can be assumed to be the same as that for repetitive reading from the disk’s buffer for the DMA test.
Using WinXP, default operation Properties Policy is Optimise For Quick Removal (OQ below). Unlike with Flash Memory USB devices, the alternative Optimise For Performance (OP) is available for USB disks. Using OP, writing and reading speeds are very similar but writing speed is faster than OQ, due to overheads. Assuming a constant data transfer speed, overheads per block transferred can be calculated, as shown by the minimum values in the table. The larger program block sizes have much higher overhead, as though transferred in multiple blocks of 32 KB to 64 KB. CPU utilisation figures have not been shown as, in most cases, they were very small. The year of purchase of the PCs is shown. Comparing these and reducing overheads, indicates that the USB controller speed might determine the penalty of older technology.
Results are now included for 64-Bit Vista, showing that the 64 bit version of the program produces the same performance as the 32 bit variety and writing speeds are slower than via XP x64.
The last two results are for a 2.5 inch 5400 RPM disk in a tiny enclosure that also obtains power from the USB. The second one is for NTFS formatting where writing small block sizes is faster and similar to results using Optimise For Performance. With FAT, writing speed is similar to the 7200 RPM disk but reading can be somewhat slower.
|
CPU MBytes/Second at KB Block Size
Key MHz Mode 1 2 4 8 16 32 64 128 256 512 1024 Max
Win2K - 2001 card Acomdata box, WD disk
yDiskUSB2d 1533 Write 0.3 0.6 1.3 2.1 3.7 6 8 9 9 9 9 9
yDiskUSB2d 1533 Read 0.6 1.2 2.4 4.1 6.2 10 13 13 13 13 13 13
yDiskUSB2d 1533 DMA 0.7 1.3 2.6 3.6 7.8 10 13 13 8 13 13 13
WinXP - 2001 card - Freecom disk - Write O/H 3.1 ms
yDiskUSB2a OQ 1900 Write 0.3 0.6 1.3 2.5 4.3 7 10 11 12 13 13 13
yDiskUSB2a OQ 1900 Read 0.7 1.3 2.6 5.2 7.7 10 14 14 14 14 14 14
yDiskUSB2a OQ 1900 DMA 0.7 1.3 2.6 5.1 7.8 10 14 14 14 14 14 14
WinXP - 2001 card - Freecom disk - Write O/H 1.5 ms
yDiskUSB2a OP 1900 Write 0.7 1.3 2.6 5.1 7.7 10 14 14 14 14 14 14
yDiskUSB2a OP 1900 Read 0.7 1.3 2.6 5.2 7.8 10 14 14 14 14 14 14
yDiskUSB2a OP 1900 DMA 0.6 1.2 2.6 5.2 7.8 10 14 14 14 14 14 14
Win2K - 2004 on-board - Freecom disk - Write O/H 0.5 ms
yDiskUSB2c OP 2080 Write 2.0 4.4 7.7 12 18 23 26 26 26 26 26 26
yDiskUSB2c OP 2080 Read 2.0 4.0 6.9 12 18 23 28 28 28 28 28 28
yDiskUSB2c OP 2080 DMA 2.0 3.9 7.5 12 18 23 28 28 28 28 28 28
WinXP x64 - 2005 on-board - Freecom disk - Write O/H 1.0 ms
yDiskUSB2b OQ 2210 Write 1.0 1.9 3.8 7 12 19 25 29 30 32 30 32
yDiskUSB2b OQ 2210 Read 2.0 3.9 7.7 13 18 27 34 34 34 34 33 34
yDiskUSB2b OQ 2210 DMA 2.0 3.2 7.8 13 16 25 31 32 33 33 33 33
WinXP x64 - 2005 on-board - Freecom disk - Write O/H 0.5 ms
yDiskUSB2b OP 2210 Write 1.9 3.9 7.7 12 20 28 34 33 33 33 34 34
yDiskUSB2b OP 2210 Read 2.0 3.9 7.9 13 18 27 33 33 35 35 35 35
yDiskUSB2b OP 2210 DMA 1.6 3.9 6.4 13 18 27 32 32 33 33 33 33
64-Bit Vista - 2007 on-board - Freecom disk - Write O/H 1.0 ms
yDiskUSB2f OQ 2400 Write 0.9 1.8 3.7 7 11 16 21 22 24 25 25 25
yDiskUSB2f OQ 2400 Read 2.0 3.9 7.8 12 18 25 33 33 33 33 33 33
yDiskUSB2f OQ 2400 DMA 2.0 3.9 7.7 13 18 26 30 33 33 33 33 33
64-Bit Vista - 2007 on-board - Freecom disk - Write O/H 0.5 ms
yDiskUSB2f OP 2400 Write 1.9 3.8 7.6 12 18 22 26 26 26 26 26 26
yDiskUSB2f OP 2400 Read 2.0 3.9 7.8 13 18 25 33 33 33 33 33 33
yDiskUSB2f OP 2400 DMA 2.0 3.9 7.7 13 18 26 33 33 33 33 33 33
64-Bit Vista - 2007 on-board - Freecom disk - 64 Bit Version
yDiskUSB2f OQ 2400 Write 1.0 1.9 3.7 7 10 16 21 23 24 25 25 25
yDiskUSB2f OQ 2400 Read 2.0 3.9 7.7 12 18 25 33 33 33 33 33 33
yDiskUSB2f OQ 2400 DMA 1.9 3.9 7.8 13 18 26 33 33 33 33 33 33
64-Bit Vista - 2007 on-board - Freecom disk - 64 Bit Version
yDiskUSB2f OP 2400 Write 1.9 3.8 7.6 12 17 22 26 26 26 26 26 26
yDiskUSB2f OP 2400 Read 2.0 3.9 7.8 13 18 25 33 33 33 33 33 33
yDiskUSB2f OP 2400 DMA 1.9 3.9 7.8 13 18 26 33 33 33 33 33 33
64-Bit Vista SP1 - 2008 on-board - Freecom disk - 64 Bit Version
yDiskUSB2f OP 2400 Write 1.9 3.8 6.2 10 15 21 25 25 26 26 26 26
yDiskUSB2f OP 2400 Read 1.9 3.9 6.2 12 18 24 33 33 33 33 33 33
yDiskUSB2f OP 2400 DMA 1.9 3.9 6.3 12 18 25 33 33 33 33 33 33
64-Bit Vista SP1 - on-board - FAT - WD Passport Disk 2009
yDiskUSB2g OQ 2400 Write 0.89 1.8 2.8 5.3 9.3 14 19 22 24 20 17 24
yDiskUSB2g OQ 2400 Read 2.5 4.6 6.9 11 17 21 26 26 25 26 25 26
yDiskUSB2g OQ 2400 DMA 2.5 5.0 7.8 14 21 28 33 33 33 33 33 33
64-Bit Vista SP1 - on-board - NTFS - WD Passport Disk 2009
yDiskUSB2g OQ 2400 Write 2.5 3.8 7.5 12 17 21 25 23 24 24 24 25
yDiskUSB2g OQ 2400 Read 2.5 4.7 7.7 14 20 27 33 31 31 33 31 33
yDiskUSB2g OQ 2400 DMA 2.5 5.0 7.8 13 21 28 33 33 33 33 33 33
|
To Start Results/Examples Index
USB and Firewire Disk Drives
The disk used for these tests is a 120 GB WD 1200JB, IDE, 7200 RPM capable or running at 748 Mbits/sec and mainly in a Hiyatec enclosure with USB and Firewire connections. Other results are via a Sohouse IDE/USB adapter and some for an earlier system (different disk).
Earlier Systems - Limited results for a PC from 2003 showed 16/24 MB/second writing/reading speed for USB 2, 19/33 MB/second for Firewire and DMA reading speed results shown below, where Firewire is again faster. Next details are for a 2003 Laptop where USB 1 is shown to be significantly slower than Firewire and the latter as fast as some modern systems.
Same Disk, Same PC, Different Interfaces - XP x64 2210 MHz has two Firewire sockets. FW1 is on a Creative sound card and FW2 from the Motherboard, the latter being up to 15% faster on writing and mainly 28% faster on reading. Then there are the Hiyatek and Sohouse USB interfaces on the 1830 MHz laptop and 2400 MHz desktop. Here, Hiyatek is up to 50% faster. Best gains tend to be at the largest block sizes.
Same Disk, Same Adapter, Different PCs - There are significant differences in speeds on the three PCs tested (like up to 35% on writing) and it is not all down to CPU MHz.
Same Disk, Same PC, Same Adapter, USB and Firewire Connections - The best Firewire speed advantages are on the 1830 MHz laptop, more than twice as fast as USB at small block sizes, 10% faster on reading with large block sizes and similar speed writing large blocks. The 2210 MHz desktop shows the same gains on reading but lower gains on writing small blocks with USB 20% faster writing with large block sizes.
Other observations to note are faster maximum DMA reading speeds of 39 to 40 MB/second with Firewire and lower CPU utilisation for both writing and reading.
|
All Results FAT Format, Optimise For Quick Removal
CPU MBytes/Second at KB Block Size
MHz Mode 1 2 4 8 16 32 64 128 256 512 1024 Max
USB v Firewire - WinXP 2003
yDiskUSB2e 3000 DMA 1.4 2.7 5.3 9 14 19 24 24 25 23 25 25
yDiskFwire 3000 DMA 2.3 4.4 8.0 14 20 27 33 37 38 39 30 39
Laptop - WinXP 2003
Hiyatek 1800 Write 0.93 4.0 6.4 10 15 20 24 25 26 27 27 27
XP FW 1800 Read 5.8 10 15 17 24 32 33 33 33 33 34 34
AMD CPU 1800 DMA 5.6 11 17 24 29 35 37 38 39 20 30 39
Hiyatek 1800 Write 0.1 0.2 0.3 0.5 0.6 0.8 0.9 0.9 0.9 0.9 0.9 0.9
XP USB 1 1800 Read 0.3 0.5 0.7 0.8 0.9 1.0 1.0 1.0 1.0 1.0 1.0 1.0
AMD CPU 1800 DMA 0.3 0.5 0.7 0.8 0.9 1.0 1.0 1.0 1.0 1.0 1.0 1.0
Same Disk, Various Systems
Hiyatek 2210 Write 0.96 3.3 6.0 9.4 14 18 21 22 23 23 23 23
XP x64 FW1 2210 Read 5.8 10 13 17 17 25 26 27 27 27 27 27
AMD dualcore 2210 DMA 6.4 11 16 20 24 26 27 28 28 20 24 28
Hiyatek 2210 Write 0.97 3.8 6.7 11 15 20 23 24 25 25 25 25
XP x64 FW2 2210 Read 6.4 10 17 18 25 33 33 35 34 34 34 35
AMD dualcore 2210 DMA 7.3 12 19 26 32 36 39 40 40 20 30 40
Hiyatek 2210 Write 0.94 2.2 3.7 7.3 12 18 26 31 32 33 33 33
XP x64 USB 2210 Read 2.5 4.8 7.3 13 17 27 31 32 32 32 32 32
AMD dualcore 2210 DMA 2.6 5.2 7.8 17 21 28 34 35 15 30 28 35
Hiyatek 1830 Write 0.94 3.8 5.9 9.7 14 18 21 23 24 24 24 24
Vista 32 FW 1830 Read 5.7 8.7 13 17 24 31 33 31 34 35 35 35
Core 2 Duo Mob 1830 DMA 4.6 8.7 16 4.8 30 33 35 39 39 20 30 39
Hiyatek 1830 Write 0.78 1.4 2.8 5.2 9.0 14 19 22 24 24 24 24
Vista 32 USB 1830 Read 2.1 3.2 6.3 10 17 25 30 31 31 31 32 32
Core 2 Duo Mob 1830 DMA 1.9 3.3 6.8 10 9.1 26 32 32 32 20 29 32
Hiyatek 2400 Write 0.95 1.9 3.5 6.3 11 16 21 23 24 25 24 25
Vista 64 USB 2400 Read 2.4 4.7 7.2 13 17 27 31 31 32 31 32 32
Core 2 Duo 2400 DMA 2.6 4.1 7.8 4.7 16 28 33 33 33 20 29 33
Sohouse 1830 Write 0.69 1.3 2.4 4.2 7.0 10 14 15 16 16 16 16
Vista 32 USB 1830 Read 1.6 3.1 5.8 8.5 13 18 21 21 22 22 22 22
Core 2 Duo 1830 DMA 1.5 3.2 6.1 8.9 14 18 22 22 22 20 20 22
Sohouse 2400 Write 0.95 1.6 3.0 5.3 8.6 12 15 16 17 17 17 17
Vista 64 USB 2400 Read 2.5 4.7 7.3 10 15 20 23 23 23 23 23 23
Core 2 Duo 2400 DMA 2.6 5.1 7.8 13 14 21 24 24 24 20 20 24
|
To Start Results/Examples Index
Local Area Network Speed
Following are examples of performance of 10 and 100 Mbits/second Local Area Networks. These were via an older version of the benchmark which, as indicated elsewhere, was slow on writing. So, an example of writing speed via the latest version is shown. DMA speeds were generally the same as reading speeds, except for the example (Win98) shown, where data is cached in the local system’s memory. Measured CPU utilisation on LAN tests can be greater than 30% on the 1900 MHz CPU but it might indicate polling time that could be used by other processes, if needed.
|
CPU Speed MBytes/Second at KB Block Size
Key MHz mbps 1 2 4 8 16 32 64 128 256 512 1024 Max
Write
zLAN10mbs LAN 450 10 0.72 0.72 0.72 0.73 0.85 0.92 0.96 0.97 0.96 0.96 0.94 0.97
zLAN100m1 LAN 1900 100 0.90 1.5 2.5 2.9 4.7 6.0 6.5 6.7 6.8 6.8 6.2 6.8
zLAN100m2 LAN 750 100 0.66 1.1 1.7 1.9 3.0 4.8 6.2 6.5 6.4 6.4 6.5 6.5
zLAN100m3 LAN 1900 100 0.68 1.1 1.7 1.9 3.1 4.6 6.2 6.4 6.3 6.2 6.5 6.5
Later 1900 100 1.1 2.1 2.9 5.6 6.9 7.5 8.5 9.2 10.2 10.4 10.5 10.5
zWiFi56 1830 56 0.6 0.9 1.4 1.8 2.1 2.3 2.5 2.6 2.7 2.8 2.8 2.8
Read
zLAN10mbs LAN 450 10 0.70 0.68 0.71 0.8 0.9 1.0 1.0 1.0 1.0 1.0 1.0 1.0
zLAN100m1 LAN 1900 100 3.2 4.7 6.7 7.7 8.7 9.3 9.4 9.6 9.7 9.8 9.8 9.8
zLAN100m2 LAN 750 100 3.4 4.6 7.0 8.0 9.2 9.9 10.0 10.2 10.4 10.5 10.5 10.5
zLAN100m3 LAN 1900 100 3.4 4.6 7.0 8.0 9.2 9.9 10.0 10.2 10.4 10.5 10.5 10.5
zWiFi56 1830 56 0.5 0.8 1.4 1.6 2.2 2.5 2.7 3.0 3.0 3.0 2.9 3.0
DMA
zLAN10mbs LAN 450 10 54 104 196 245 309 363 87 136 131 135 126 363
zWiFi56 1830 56 0.6 0.9 1.6 1.9 2.4 2.5 2.8 2.9 2.9 3.0 3.0 3.0
|
To Start Results/Examples Index
USB, Compact Flash and Secure Digital
The results for this benchmark are for old Compact Flash Memory cards which restrict the maximum reading speed. The following show speeds via a reader on an LP Port, PCMCIA on laptops and different memory cards via USB 1 and High Speed USB 2. See CDDVDSpd Results.htm for a more appropriate benchmark with results on faster cards.
The last results shown are for a USB1 to USB1 LAN type communications box.
In order to demonstrate similar characteristics, new results have been added for SD cards used with a camera. The speeds can be compared with those for 4 GB Flash Drives below.
|
CPU Device MBytes/Second at KB Block Size
Key MHz 1 2 4 8 16 32 64 128 256 512 1024 Max
Write
zCFlashLP 950 LP Port 0.15 0.18 0.21 0.23 0.23 0.24 0.26 0.27 0.27
zCFPcmcia 400 Pcmcia 0.28 0.35 0.40 0.42 0.44 0.45 0.45 0.45 0.45 0.45 0.46 0.46
zCFPcmci2 1800 Pcmcia 0.07 0.13 0.20 0.28 0.36 0.41 0.45 0.47 0.48 0.49 0.50 0.50
zUSB1CF 1900 USB1 0.05 0.08 0.13 0.21 0.29 0.36 0.40 0.43 0.44 0.45 0.45 0.45
zUSB1CF 1900 USB1 0.03 0.05 0.09 0.16 0.24 0.32 0.39 0.44 0.44 0.46 0.50 0.50
zUSB2CF 1900 USB2 0.05 0.09 0.14 0.22 0.30 0.37 0.42 0.44 0.46 0.47 0.47 0.47
zUSB2CF 1900 USB2 0.03 0.06 0.11 0.18 0.27 0.39 0.45 0.50 0.54 0.55 0.62 0.62
zUSBComms 1900 USB1 0.12 0.15 0.25 0.39 0.42 0.50 0.54 0.57 0.57 0.59 0.59
zSD32MB 2400 USB2 0.03 0.06 0.12 0.25 0.52 0.73 0.89 1.00 1.10 1.10 1.20 1.20
zSD2GB 2400 USB2 0.13 0.27 0.52 0.98 1.90 3.40 5.40 7.70 9.70 9.30 8.90 9.70
Read
zCFlashLP 950 LP Port 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1
zCFPcmcia 400 Pcmcia 0.7 0.9 1.1 1.2 1.2 1.3 1.3 1.3 1.3 1.3 1.3 1.3
zCFPcmci2 1800 Pcmcia 0.8 1.1 1.3 1.5 1.6 1.6 1.6 1.6 1.7 1.7 1.6 1.7
zUSB1CF 1900 USB1 0.1 0.2 0.4 0.6 0.7 0.8 0.9 0.9 0.9 0.9 0.9 0.9
zUSB1CF 1900 USB1 0.2 0.3 0.4 0.5 0.7 0.8 0.8 0.8 0.8 0.8 0.8 0.8
zUSB2CF 1900 USB2 0.3 0.5 0.8 1.1 1.3 1.4 1.5 1.5 1.5 1.5 1.5 1.5
zUSB2CF 1900 USB2 0.6 1.0 1.4 1.6 1.8 1.9 1.9 1.9 1.9 1.9 1.9 1.9
zUSBComms 1900 USB1 0.1 0.2 0.3 0.4 0.4 0.4 0.5 0.5 0.5 0.5 0.5
zSD32MB 2400 USB2 1.1 1.7 2.3 2.8 3.2 3.4 3.6 3.6 3.6 3.5 3.6 3.6
zSD2GB 2400 USB2 1.1 2.0 3.6 6.1 8.9 11.8 16.9 16.9 17.0 16.9 17.0 17.0
|
To Start Results/Examples Index
4 GB USB Flash Drives
The benchmark was run on two 4 GB Flash Drives with FAT and NTFS formatting, using both 32 bit and 64 bit benchmarks via Windows XP Pro x64 and 64-Bit Vista, plus the 32 bit version using Windows 2000 (Systems see Key jC2Dsea2, q64AMax1 and oThoHi1).
Results via Vista and x64 were essentially the same as were those for the 32 bit and 64 bit benchmarks. So, the following are example results and some of these show the same performance. The Windows 2000 system is the only one where the optimise for performance setting works. This leads to faster writing speeds with FAT formatting and small block sizes. NTFS has the same impact. The second flash drive is clearly much faster, except reading with small block sizes.
The benchmark was also run using 1 GB file size to measure random access times (see below). These were quite consistent on the second drive.
The first drive was found unsuitable for Vista ReadyBoost, reported as insufficient write performance at 1723 KB/sec. The minimum requirements for ReadyBoost are 2.5 MB/sec for 4 KB random reads and 1.75 MB/sec for 512K random writes. The second drive, with NTFS format, was reported as reading at 5486 KB/sec and writing at 1814 KB/sec.
The reading speed makes sense, compared with these benchmark results. Random reading of 1 KB at 0.52 milliseconds is at a speed of 1.92 MB/sec, virtually the same as that for the benchmark reading 1 KB blocks, with 5.4 MB/sec at 4 KB. On the other hand, it is difficult to see where the 1814 KB/sec comes from (particularly if it is really for 512 KB) with this benchmark indicating greater than 3 MB/sec. My other benchmark (see CDDVDSpd Results.htm) shows that individual small files can be written faster than the ReadyBoost measurements.
I suppose that one way to achieve the low speeds would be to produce a fragmented data space where writing 512 KB fills in the spaces, effectively as small block sizes.
|
MBytes/Second at KB Block Size
Key Win 1 2 4 8 16 32 64 128 256 512 1024
Write
zFD4096A FAT 2K 0.21 0.37 0.46 0.52 0.59 2.60 2.60 2.60 2.60 2.60 2.50
zFD4096A FAT 64 0.07 0.10 0.14 0.24 0.30 1.80 2.40 2.60 2.60 2.50 2.40
zFD4096A NTFS 64 0.21 0.37 0.47 0.54 0.59 2.50 2.50 2.70 2.70 2.60 2.60
zFD4096C FAT 2K 0.62 0.97 1.40 2.20 2.90 3.80 4.40 4.50 4.50 4.30 4.40
zFD4096C FAT 64 0.17 0.32 0.56 1.00 1.80 2.70 3.60 4.10 4.40 4.50 4.50
zFD4096C NTFS 64 0.61 0.95 1.40 2.30 3.40 4.20 4.80 4.00 3.90 3.60 3.30
Read
zFD4096A FAT 2K 2.20 3.80 5.20 6.80 7.70 8.60 9.10 9.20 9.20 9.20 9.20
zFD4096A FAT 64 1.90 3.90 5.20 6.90 7.80 8.60 9.10 9.10 9.20 9.20 9.20
zFD4096A NTFS 64 1.90 3.90 5.20 6.80 7.80 8.70 9.10 9.20 9.10 9.20 9.20
zFD4096C FAT 2K 1.90 3.10 5.40 8.80 11.40 14.60 16.60 16.60 16.60 16.60 16.60
zFD4096C FAT 64 1.90 3.10 5.40 8.90 12.40 14.60 16.50 16.30 16.30 16.80 16.60
zFD4096C NTFS 64 1.90 3.10 5.40 8.90 12.30 14.60 16.50 16.60 16.50 16.60 16.50
Random reading milliseconds 1 KB from 1 MB to 1024 MB data
zFD4096A FAT 64 0.51 0.50 0.50 0.51 0.51 0.50 0.50 0.50 0.89 1.61 2.39
zFD4096A NTFS 64 0.65 0.54 1.31 1.70 1.27 1.17 0.84 0.71 0.59 2.18 2.22
zFD4096C FAT 64 0.51 0.52 0.52 0.51 0.52 0.53 0.54 0.57 0.54 0.61 0.61
zFD4096C NTFS 64 0.52 0.52 0.56 0.55 0.52 0.53 0.54 0.58 0.60 0.62 0.66
|
To Start Results/Examples Index
DMA Problems
Following are three examples of DMA found not to be working on running the benchmark and results after enabling DMA. The slow PIO mode is identified by CPU utilisation approaching 100%. The results suggest that the relative performance loss without DMA has become progressively worse. The fourth example shows variations on the same PC when booting to Windows 98, which uses UDMA33, and Windows XP at UDMA100.
Disk reading speed is shown but there were similar differences on writing.
|
|
CPU Disk MBytes/Second at KB Block Size % CPU Utilisation
Key MHz RPM 1 2 4 8 16 32 64 128 256 512 1024 1 2 4 8 16 32 64
K63WD 400 5400 1.4 1.4 1.6 2.4 2.7 3.9 4.3 3.3 3.7 3.8 4.0 87 87 87 79 75 75 66
K63WD DMA 400 5400 4.4 5.8 5.9 6.0 6.1 6.3 6.9 8.0 8.0 8.1 8.0 65 72 66 64 63 65 69
eCelMax1a 950 3.3 3.9 4.4 4.6 4.8 4.8 5.0 5.0 5.0 5.0 5.0 80 92 93 95 97 93 93
eCelMax1a DMA 950 7.5 12 18 23 26 28 29 30 30 30 30 54 41 26 17 10 5 2
gP3IBM1 1000 7200 2.1 2.3 2.3 2.3 2.4 2.3 2.3 2.3 2.4 2.4 2.4 96 96 94 95 97 94 95
gP3IBM1 DMA 1000 7200 11 23 32 31 32 32 32 32 32 31 32 64 71 49 25 12 6 3
Win98 UDMA33 WinXP UDMA 100
oPalWD2 W98 1466 7200 9.3 13 16 18 21 20 20 19 20 24 26 45 29 17 9 5 2 1
oPalWD2 W2K 1466 7200 11 16 24 23 24 43 44 43 44 44 43 23 19 20 8 5 4 2
|
|
Some excessive DMA speeds can be produced when Norton Anti-Virus is in use. It seems that this software reads the first part of a file using the Windows file caching option. Then the DMA test is repetitively reading the data from memory. The following examples show this effect up to 16 KB block size.
|
|
CPU Disk MBytes/Second at KB Block Size % CPU Utilisation
Key MHz RPM 1 2 4 8 16 32 64 128 256 512 1024 1 2 4 8 16 32 64
eCelNA2 AV 900 50 93 167 234 51 56 59 58 22 23 23 100 100 100 100 76 42 29
eCelMax2 AV 700 59 105 172 286 382 50 40 34 32 20 24 100 100 100 100 100 14 7
|
To Start Results/Examples Index
Driver Problems
The following shows results of an experiment in comparing Microsoft and Intel bus drivers. Both produce a strange reduction in performance at large block sizes but in different places.
Next is an example of how poor performance was corrected by re-installing the bus driver.
|
CPU Disk MBytes/Second at KB Block Size
Key MHz RPM 1 2 4 8 16 32 64 128 256 512 1024
Write
ip4WD2 Intel 2411 7200 1.0 6 10 18 28 40 51 30 63 64 65
ip4WD2 MS 2411 7200 1.0 6 9 18 28 40 36 35 63 64 65
Read
ip4WD2 Intel 2411 7200 9.5 19 37 41 42 42 42 19 25 22 28
ip4WD2 MS 2411 7200 9.5 19 37 41 42 42 43 43 43 28 35
Driver Reloaded
nTbdMax1 Wr1 1533 5400 1.4 2.6 5.6 6.7 9.3 10 10 11 11 11 11
nTbdMax1 Wr2 1533 5400 1.5 3.0 5.4 11 19 24 26 26 27 27 21
|
|
This example demonstrates different speeds on a board where the disk can be plugged into a RAID or IDE connector.
|
CPU Disk MBytes/Second at KB Block Size
Key MHz RPM 1 2 4 8 16 32 64 128 256 512 1024
oPalIBM3 RAID 1539 7200 6.6 13 16 19 21 26 23 26 26 27 24
oPalIBM3 IDE 1539 7200 6.6 12 22 36 37 37 37 38 37 40 40
|
To Start Results/Examples Index
CD and DVD Writers
The benchmark can also be used for testing CD/DVD drives with RW discs when packet writing software is available, such as Nero InCD. Following are example results on a DVD drive.
Note the variations in speed and the high CPU utilisation.
|
Pentium 4 1.9 GHz, Liteon LH-20A1H DVD +RW 8x, DVD Disc Rating ??,
using Nero InCD, WinXP
###########################################################
Disk Test DiskGraf Version 1.3A Sun Jun 24 16:33:49 2007
Copyright Roy Longbottom 2001
Windows NT Version 5.1, build 2600, Service Pack 1
CPU GenuineIntel, Features Code 3FEBFBFF, Model Code 00000F12, 1900 MHz
Memory 523760 KB, Free 250432 KB
F:Disk/partition 4479 MB, Free 4478 MB
Files 5 Size 8 MB Block Sizes 10 Minimum RAM Data
Block File ->
KB 1 2 3 4 5
Write MB/Second
2 2.1 3.4 1.7 3.4 3.4
4 2.1 1.7 3.4 3.4 3.4
8 1.3 3.4 3.4 3.4 1.8
16 2.2 3.4 3.4 1.8 3.4
32 2.2 3.4 1.6 3.4 3.4
64 2.0 1.8 3.4 3.4 3.4
128 1.3 3.4 3.4 3.4 1.7
256 2.0 3.4 3.4 1.7 3.4
512 2.1 3.4 1.7 3.4 3.4
1024 2.0 1.7 3.3 3.3 3.3
Read MB/Second
2 2.6 3.2 3.2 2.6 5.1
4 2.6 3.2 3.2 2.6 5.1
8 2.6 3.2 3.2 2.6 5.1
16 2.6 3.2 3.2 2.6 5.1
32 2.5 3.2 3.2 2.7 5.1
64 2.6 3.2 3.2 2.7 5.1
128 2.6 3.2 3.3 2.6 5.2
256 2.6 3.2 3.2 2.7 5.2
512 2.6 3.2 3.3 2.7 5.2
1024 2.6 3.2 3.2 2.7 5.1
Write CPU Utilisation
2 55.1 85.2 45.1 85.0 84.6
4 32.5 28.0 47.7 48.1 47.9
8 12.0 28.7 28.7 28.4 18.4
16 13.2 20.0 20.5 12.1 20.2
32 10.4 15.9 8.6 15.6 15.4
64 7.6 9.3 13.3 12.7 13.3
128 4.8 11.4 11.4 11.7 8.0
256 5.5 10.3 10.5 5.9 10.9
512 5.7 9.4 5.4 9.8 9.5
1024 5.9 5.0 7.8 8.2 8.5
Read CPU Utilisation
2 41.0 47.6 47.4 38.2 74.1
4 35.1 40.8 39.9 33.3 63.7
8 29.9 34.3 34.0 27.4 53.8
16 26.0 30.6 30.9 26.3 48.6
32 24.1 28.7 28.5 23.6 46.4
64 23.9 27.5 27.5 22.7 44.5
128 23.4 26.5 26.4 21.7 43.8
256 23.3 29.4 27.1 22.3 43.7
512 21.5 26.5 26.1 21.5 42.8
1024 21.7 26.1 26.6 21.2 42.4
DMA Reread block up to 100 times at 1 to 1024 Kbytes
Read MB/Second
3.1 10.5 24.7 39.9 56.7 72.7 75.4 72.1 75.1 74.6 76.4
Read CPU Utilisation
100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Random Read 1KB out of 1, 2, 4, to file size Mbytes
Read Milliseconds
5.86 19.6 40.3 14.9 0.000 0.000 0.000 0.000 0.000 0.000 0.000
Read CPU Utilisation
6.0 2.3 2.5 2.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0
End Disk Test Sun Jun 24 16:39:13 2007
|
To Start Results/Examples Index
|
Disk Results
The following results include some for the 64 bit version as well at those at 32 bits - q64AMax1 via Windows XP Pro x64 and jC2DSea2 via 64 Bit Widows Vista. Results for the latter are also provided for the two 200 GB partitions C: and D:. A surprise is that speed on the smaller block sizes is much slower on the first partition (C:). The same was apparent on a PC with Windows XP. Perhaps there is a Windows overhead checking data transfers on the System partition (there was no Anti-Virus software in use).
Writing Speed
CPU Disk MBytes/Second at KB Block Size % CPU Utilisation
Key MHz RPM 1 2 4 8 16 32 64 128 256 512 1024 Max 1 2 4 8 16 32 64
a486DX2 66 3600 0.35 0.38 0.54 0.71 0.94 1.1 0.94 0.97 1.1 1 22 19 26 33 44
bPenSea 100 5400 0.95 1.2 1.5 1.5 1.6 1.7 1.7 1.8 1.7 1.8 1.5 2 84 82 87 87 90 94 95
dPProQua # 200 4500 0.17 0.33 0.59 1.0 1.5 2.1 2.7 3.6 4 4 6 10 16
fPIILap 300 5400 2.5 4.8 5.4 5.4 5.4 5.2 5.4 5.4 5.2 5.4 5.4 5 23 22 13 8 4 3 2
dPProQua 200 4500 1.2 2.3 4.1 4.8 5.4 5.8 5.8 5.8 5.8 5.3 5.3 6 36 52 77 83 88 90 88
eCelLap 400 4200 2.5 3.6 4.1 4.8 5.1 5.1 5.2 5.4 5.4 6.1 6.1 6 23 14 9 5 3 2 2
K62Sea2 380 5400 2.4 2.9 3.5 3.4 3.4 3.5 3.5 3.5 3.5 5.6 6.3 6 29 18 12 8 7 6 5
cPMMXQua 200 5400 1.3 2.5 4.8 7.4 7.3 7.3 7.4 7.4 7.1 7.1 7.1 7 29 42 50 81 74 71 72
gP3Max # 1400 1.2 2.4 4.1 6.3 7.3 9.4 9.6 9.8 11 11 11 11 18 16 13 11 7 5 2
nTbdQuaRd IR# 1400 7200 0.15 0.27 0.47 0.90 1.6 3.1 5.0 9.0 11 11 12 12 12 12 12 13 13 13 13
pBarLap # 1800 4200 0.68 1.2 2.2 3.6 5.8 8.1 8.7 3.1 5.3 8.4 12 12 6 5 5 4 3 2 0
eCelMax 450 7200 2.4 4.8 7.8 11 9.2 12 12 9 12 12 12 12 21 22 15 12 5 4 2
fPIIFuj 450 5400 2.5 4.7 6.7 11 12 12 12 12 12 12 12 12 22 21 15 12 7 4 2
oPalMaxRd $IR# 1667 7200 0.2 0.3 0.6 1.2 2.0 3.1 6.7 8.4 12 12 D
gPIIILap 650 7.6 10 12 12 13 13 12 13 12 12 12 13 41 32 19 11 6 3 2
iP4Sea2 # 1900 5400 1.1 1.8 2.7 4.3 7.0 10 11 13 13 15 15 15 16 14 14 11 7 4 1
K62Sea # 500 7200 1.1 2.1 3.1 6.1 5.7 6.6 6.5 10 14 15 17 17 63 58 46 33 20 11 7
gp3SeaA # 1400 7200 1.3 2.4 4.2 5.5 8.5 12 13 16 16 16 17 17 15 14 10 9 7 4 3
fPIISea1 SC 450 10000 7.9 14 17 17 18 17 17 17 17 18 17 18 74 58 39 22 14 8 6
iP4NLap # 1700 5400 0.75 1.4 2.8 4.7 9.0 15 18 5 8 12 16 18 8 8 8 7 6 5 1
iP4Sam # 3066 5400 0.52 1.1 1.9 3.9 8.6 12 14 14 18 16 14 18 3 2 2 3 2 1 1
eCelMax2 # 700 1.7 3.0 5.0 7.2 8.9 10 11 12 17 18 17 18 30 27 25 16 10 6 3
K63WD 400 5400 0.18 0.36 15 16 16 16 16 16 16 16 18 18 1 1 20 13 8 6 6
oPalSea3 SC# 1667 15000 0.66 1.4 2.5 4.9 10 8.9 14 16 18 18 D
oPalIBM2 # 1466 7200 0.67 1.6 3.0 5.8 9.4 13 17 16 18 18 16 18 6 7 7 7 5 3 2
eCelMax1 950 5400 6.6 11 14 16 18 18 19 18 19 19 19 19 36 31 20 13 8 5 4
oPalMax1 # 1667 7200 1.1 2.3 3.2 5.1 10 18 18 17 16 17 19 19 D
eCelMax2 700 7.4 11 15 17 18 19 19 19 19 19 19 19 40 35 23 16 10 7 6
iP4Sea2 1900 5400 3.6 6.4 10 18 20 21 21 21 21 21 21 21 24 20 17 15 7 4 1
eCelNA2 900 5.3 9.9 16 20 22 22 22 22 22 22 22 22 58 55 46 31 19 13 9
oPalQuaS SC# 1533 10000 0.26 0.49 1.1 1.8 4.1 6.2 10 16 22 22 D
mDurMax2 950 7200 4.2 8 11 18 22 24 23 24 24 24 24 24 36 37 24 20 13 8 9
iP4Sea3 # 2533 7200 2.1 3.9 6.2 8.9 12 18 23 23 22 23 27 27 16 17 15 10 6 3 1
nTbdMax1 # 1533 5400 1.5 3.0 5.4 11 19 24 26 26 27 27 21 27 10 10 9 8 8 3 1
iP4Sea1 # 2016 7200 1.9 3.1 4.4 7.6 13 19 20 22 21 22 27 27 30 36 26 27 22 15 7
oPalWD1 # 1410 7200 1.4 5.5 11 18 26 25 23 27 28 27 28 28 19 29 45 39 27 14 6
oPalIBM3 1539 7200 6.8 13 18 22 25 27 28 28 28 28 28 28 41 37 27 15 9 5 2
nTbdWDI1 # 1300 7200 1.7 6.0 13 20 28 26 25 29 29 28 28 29 17 39 43 30 22 10 5
eCelWD1 466 7200 0.96 10 17 21 24 27 28 28 29 30 29 30 11 52 42 28 21 15 11
oPalSea2 %SC# 1700 7200 2.0 4.0 7 11 12 18 20 19 23 23 30 30 19 23 20 14 6 5 1
iP4Xeon # 2175 1.8 3.6 6.3 10 14 22 24 24 25 24 30 30 D
iP4Sea4 # 2533 7200 1.7 2.1 3.9 6.7 14 21 25 28 31 33 31 33 14 9 6 5 5 4 2
q64ASea1 # 2150 7200 1.8 2.5 3.7 8.2 14 16 21 27 30 33 34 34 8 6 4 5 4 2 1
nTbdQua 1200 7200 3.8 7.5 15 29 36 36 35 35 35 35 35 36 14 14 14 14 6 4 2
gP3Sea15 %SC# 667 15000 0.23 0.45 0.88 1.7 3.6 5.7 11 12 18 27 36 36 D
gP3WD1 # 1000 7200 1.7 5.9 14 24 35 34 32 36 34 37 36 37 12 30 36 25 27 11 6
iP4EWD1 SA 3000 7200 8.0 13 16 19 22 24 24 26 30 33 37 37 11 10 5 3 1 1 0
iP4WD3 # 2533 7200 0.75 4.1 12 21 31 34 34 34 34 37 38 38 6 16 17 27 19 8 3
nTbdIBM8 # 1400 7200 1.0 2.3 5.0 10 17 19 22 24 33 38 30 38 15 17 16 19 14 8 5
q64TLap D:SA 1900 5400 8.6 15 25 32 34 38 36 37 37 37 33 38 D
gP3RAID IR# 1000 0.95 1.8 3.5 6.2 11 18 24 29 35 39 39 39 D
oThoWD3 # 2300 7200 0.78 5.6 12 20 33 32 35 38 39 39 38 39 5 13 18 10 14 7 3
oPalWD2 # 1466 7200 0.86 6.8 13 24 35 34 36 40 39 40 38 40 8 23 24 17 20 10 5
gP3IBM1 1000 7200 7.6 15 27 41 41 41 41 41 41 41 41 41 49 48 46 26 8 5 3
jC2CLap SA 2000 5400 3 4 7 10 20 24 34 34 44 37 34 44 18 17 18 16 8 5 3
oPalSeaR0 %SC# 1700 15000 1.4 2.4 3.5 6.1 10 15 25 35 42 45 45 45 15 14 14 11 8 5 4
oPalSeaRd %SR# 1700 15000 1.6 2.9 5.4 9.0 15 22 30 35 41 44 46 46 16 15 14 11 9 6 4
iP4Hit1 1900 7200 6.2 12 21 36 43 44 42 47 45 44 44 47 44 42 40 32 14 8 3
jC2DLap SA 1830 5400 4 8 15 26 44 42 47 45 44 29 44 47 D
jC2DSea2 C:SA 2400 7200 6 8 14 17 25 35 38 42 46 46 52 52 D
jC2DSea2 C:SA 2400 7200 5 8 10 16 24 30 19 30 42 49 35 49 D&14 9 6 4 1 1 0
jC2DSea2 D:SA 2400 7200 10 20 40 47 47 47 52 51 51 50 49 52 D
jC2DSea2 D:SA 2400 7200 9 16 31 36 38 45 45 44 45 44 43 45 D&21 19 13 10 0 2 0
q64ASeaR SR# 1991 15000 0.33 0.62 1.2 2.4 4.3 7.3 13 21 29 32 52 52 D
oThoHi1 2080 7200 6.5 12 22 36 45 52 47 46 47 47 46 52 16 15 14 10 5 2 1
iP4Max 2400 5400 10 21 38 51 52 52 52 52 52 49 53 53 45 47 37 27 10 4 1
K62IBM 267 7200 5.8 11 19 28 38 46 50 46 52 53 54 54 85 91 91 84 66 50 40
nTbdIBMRd IR# 1400 7200 1.4 2.5 3.7 7.3 12 24 34 44 46 54 52 54 15 17 13 13 11 10 8
jC2DSea1 SA 2400 7200 5.4 8.2 15 17 27 35 38 43 48 54 57 57 D
q64AMax1 &SA 2210 7200 17 29 44 46 51 55 59 55 61 58 59 61 D&23 18 11 4 3 1 1
q64AMax1 SA 2210 7200 17 29 45 58 54 61 60 54 59 61 62 62 D
oPalIBMRd $# 1396 7200 1.3 2.7 5.2 8.7 19 24 40 54 48 62 62 62 21 20 21 19 16 21 18
iP4WD1 # 2411 7200 0.88 6.2 13 25 37 47 55 60 62 63 64 64 5 23 28 27 20 13 7
ip4WD2 # 2411 7200 1.0 5.6 10 18 28 40 51 30 63 64 65 65 8 32 30 31 26 17 4
gP3IBMRd SR# 1000 10000 1.0 1.9 3.4 6.2 9.6 18 29 46 57 69 52 69 D
iP4SeaRd SR# 3633 15000 1.2 2.3 4.1 7.1 12 22 31 43 53 61 71 71 8 7 7 5 4 3 2
iP4WD4 # 3000 7200 0.94 5 15 29 50 61 75 82 85 86 84 86 4 6 9 18 16 10 7
jC2DSea3 &ES 2400 7200 9.3 18 29 46 66 82 94 99 98 100 99 100 D&32 31 23 21 17 10 9
at 145 GB &ES 2400 7200 8.4 16 28 46 67 87 99 100 99 100 98 100 D&27 25 24 20 15 10 8
at 290 GB &ES 2400 7200 9.1 16 28 45 65 67 67 67 67 68 67 68 D&30 29 24 20 14 8 5
qPheWD1 &SA 3000 5400 11 22 39 63 64 89 75 94 94 86 92 94 Q&14 14 14 15 13 10 10
at 224 GB &SA 3000 5400 12 26 45 55 85 84 89 90 88 88 87 90 Q&15 16 14 13 11 8 9
at 448 GB &SA 3000 5400 13 26 42 59 73 72 74 71 75 72 76 76 Q&16 16 15 14 10 9 7
q64ASeaR *SR 1991 15000 7.4 13 20 39 60 82 106 108 112 133 133 133 D
iP4WDSRd AR# 3633 10000 3.4 7.8 15 30 46 62 88 109 154 160 159 160 27 30 26 25 18 12 13
iP4HiSRd AR 3000 7200 5.0 10 19 35 56 69 82 110 155 166 168 168 11 12 12 11 9 4 3
jI7ISSD SA 2800 SSD 12 24 32 30 29 38 37 46 55 83 81 83 Q
q64ASeaR2 *SR 1991 15000 6.5 13 23 41 69 101 147 190 200 218 225 225 D
# = Old Version % = PCI 50MHz SA = SATA SC = SCSI AR = SATA Raid IR = IDE RAID SR = SCSI Raid D = Dual CPU
Q = Quad CPU ES = eSATA $ = High Speed PCI * = PCI-X/64bit 133 MHz & = 64 bit version C:/D: 200 GB partitions
Reading Speed
CPU Disk MBytes/Second at KB Block Size % CPU Utilisation
Key MHz RPM 1 2 4 8 16 32 64 128 256 512 1024 Max 1 2 4 8 16 32 64
a486DX2 66 3600 0.79 1.1 1.0 1.1 1.3 1.3 1.2 1.2 1.3 1 66 59 46 48 54
bPenSea 100 5400 1.2 1.3 1.3 1.6 1.6 1.7 1.7 1.7 1.7 1.7 1.4 2 97 97 98 97 95 98 99
dPProQua 200 4500 3.4 3.5 3.5 3.3 3.1 3.2 3.1 4.3 4 71 76 77 27
dPProQua 200 4500 2.6 2.7 2.9 3.3 3.3 3.8 4.4 4.4 4.3 4.2 4.2 4 70 72 70 74 73 80 93
fPIILap 300 5400 3.4 5.3 5.3 5.3 5.3 5.5 5.5 5.3 5.5 5.5 5.3 6 29 11 13 7 4 3 2
K62Sea2 380 5400 2.1 3.4 4.6 5.7 5.3 5.1 4.9 4.9 4.8 6.9 6.1 7 25 20 12 6 4 3 2
eCelLap 400 4200 3.7 5.5 6.7 6.8 6.7 6.8 6.7 6.7 6.7 7.2 7.2 7 32 23 14 7 4 2 1
cPMMXQua 200 5400 5.4 6.2 7.0 7.7 7.5 7.5 7.4 7.5 7.2 7.2 7.2 8 86 85 93 76 70 66 67
eCelMax 450 7200 6.5 9.9 12 12 12 12 12 12 12 12 12 12 47 38 23 11 6 3 2
fPIIFuj 450 5400 5.4 7.9 11 12 12 12 12 12 12 12 12 12 42 28 21 12 5 3 2
gPIIILap 650 11 15 15 15 15 15 15 15 15 15 15 15 61 41 21 10 5 3 1
K63WD 400 5400 5.8 8.3 8.5 8.8 8.9 16 16 16 16 17 16 17 33 22 12 7 4 2 2
fPIISea1 SC 450 10000 8.0 14 17 17 18 18 18 18 18 18 18 18 73 65 42 22 13 8 5
gp3SeaA 1400 7200 4.6 6.8 9.0 9.7 14 15 18 20 20 20 20 20 15 10 5 4 4 2 1
K62Sea 500 7200 4.0 6.7 9.4 13 14 16 18 20 20 19 19 20 34 40 14 19 12 6 5
pBarLap 1800 4200 17 18 19 20 21 20 21 21 21 21 21 21 42 31 17 8 3 1 0
gP3Max 1400 5.2 8.7 12 20 22 22 22 22 22 22 22 22 19 13 7 8 6 3 2
iP4Sea2 1900 5400 6.1 11 14 25 26 26 25 25 25 25 25 26 29 27 16 12 6 2 0
iP4NLap 1700 5400 12 20 26 26 26 26 26 26 26 26 26 26 77 65 39 18 8 3 0
nTbdMax1 1533 5400 11 19 19 20 21 26 27 27 27 27 18 27 10 16 1 3 1 0 0
mDurMax2 950 7200 7.9 10 14 23 27 27 27 26 26 27 26 27 60 36 27 23 13 6 8
eCelNA2 900 5.0 9.5 18 27 27 27 27 28 28 28 28 28 31 25 23 26 14 8 5
eCelMax2 700 8.9 14 19 24 27 29 30 30 30 30 30 30 49 39 26 17 10 6 4
eCelMax1a 950 7.5 12 18 23 26 28 29 30 30 30 30 30 54 41 26 17 10 5 2
oPalWD1 1410 7200 7.4 13 13 14 14 31 32 26 25 25 25 32 31 35 20 10 5 6 3
oPalMax1 1667 7200 6.9 12 10 13 17 23 29 30 32 32 31 32 D
iP4Sam 3066 5400 5.9 10 18 32 29 30 30 26 31 32 30 32 14 14 11 9 4 2 1
gP3IBM1 1000 7200 11 23 32 31 32 32 32 32 32 31 32 32 64 71 49 25 12 6 3
nTbdWDI1 1300 7200 8.3 13 13 14 14 32 33 27 16 27 25 33 36 34 17 8 4 5 2
gP3WD1 1000 7200 8.1 13 14 15 14 33 33 26 27 16 25 33 30 27 13 5 3 4 2
nTbdQua 1200 7200 13 34 34 33 35 35 34 35 34 35 34 35 47 57 30 15 7 4 2
iP4EWD1 SA 3000 7200 9.1 14 19 22 22 22 28 27 29 36 37 37 6 5 4 3 1 1 0
oPalIBM2 1466 7200 15 27 32 32 32 32 32 32 32 32 38 38 43 39 25 13 6 2 1
jC2CLap SA 2000 5400 5.4 8.9 16 18 29 33 33 36 36 38 37 38 24 19 14 9 10 6 6
iP4Xeon 2175 6.5 11 18 20 22 16 16 38 40 40 39 40 D
iP4Sea1 2016 7200 5.3 9.0 12 16 22 39 34 40 39 39 38 40 24 28 8 12 9 8 2
iP4Sea4 2533 7200 4.2 7.3 10 16 33 30 30 34 39 40 40 40 14 10 4 2 2 0 0
oPalQuaS SC 1533 10000 0.65 1.2 2.4 4.5 9.3 20 25 35 40 40 D
q64ASea1 2150 7200 6.4 11 11 22 28 26 28 32 35 38 41 41 8 10 2 5 3 0 0
iP4Sea3 2533 7200 7.3 10 12 18 33 38 40 41 40 40 40 41 27 20 11 6 6 2 0
oPalSea2 %SC 1700 7200 5.5 10 12 14 16 28 35 38 40 40 41 41 20 22 14 7 3 2 1
oPalIBM3 1539 7200 11 21 33 41 41 41 40 41 41 40 40 41 64 61 41 27 14 6 3
eCelWD1 466 7200 7.3 11 15 16 16 41 42 37 42 40 39 42 65 48 34 20 12 11 9
ip4WD2 2411 7200 9.5 19 37 41 42 42 42 19 25 22 28 42 18 25 16 0 11 8 0
iP4WD3 2533 7200 5.2 14 22 22 23 37 40 36 38 43 42 43 21 13 5 10 4 2 0
jC2DLap SA 1830 5400 5.3 10 20 29 39 41 44 40 35 33 32 44 D
oThoWD3 2300 7200 6.7 16 23 23 23 36 41 44 44 44 43 44 15 19 15 4 3 2 1
oPalWD2 1466 7200 11 16 24 23 24 43 44 43 44 44 43 44 23 19 20 8 5 4 2
gP3RAID IR 1000 8.8 16 23 30 33 39 40 41 41 45 39 45 D
iP4Max 2400 5400 15 26 41 46 46 46 46 46 46 46 46 46 62 52 42 21 9 3 0
nTbdQuaRd IR 1400 7200 1.3 1.3 1.2 1.5 3.1 5.3 11 18 28 42 46 46 16 9 6 4 5 5 6
K62IBM 267 7200 7.2 13 22 31 41 45 46 44 47 47 47 47 99 98 96 80 57 40 27
iP4WD1 2411 7200 9.9 17 38 41 43 47 46 25 25 22 30 47 25 27 24 17 13 9 7
iP4WD4 3000 7200 11 20 25 25 25 46 47 47 47 45 45 47 22 19 12 5 1 1 0
q64TLap D:SA 1900 5400 10 20 33 49 48 48 49 50 46 50 40 50 D
iP4Hit1 1900 7200 11 21 28 37 50 51 50 49 50 49 49 51 53 45 32 19 9 5 1
nTbdIBM8 1400 7200 7.1 13 22 36 46 49 51 51 51 49 49 51 25 32 10 24 15 9 5
oPalSea3 SC 1667 15000 25 34 42 49 53 54 54 54 54 54 D
gP3Sea15 $SC 667 15000 10 19 34 54 56 56 56 56 56 57 56 57 D
oThoHi1 2080 7200 12 21 32 51 57 57 57 57 57 56 57 57 27 25 18 14 8 4 1
oPalSeaR0 %SC 1700 15000 2.5 5.3 10 19 29 44 44 56 57 57 57 57 6 7 3 6 4 0 0
oThoHi1 2080 7200 11 21 32 46 53 56 60 60 60 60 59 60 26 24 18 13 7 3 1
q64AMax1 &SA 2210 7200 20 34 59 51 56 52 61 56 62 61 60 62 D&12 5 5 1 2 1 0
q64AMax1 SA 2210 7200 21 38 53 59 54 66 64 59 64 63 62 66 D
oPalIBMRd $IR 1396 7200 11 18 26 29 46 65 68 63 61 58 62 68 52 56 34 17 10 16 9
jC2DSea2 C:SA 2400 7200 7 10 14 16 30 38 42 49 55 55 59 59 D
jC2DSea2 C:SA 2400 7200 7 11 10 20 27 32 41 47 56 56 57 57 &D10 7 5 3 2 0 0
jC2DSea2 D:SA 2400 7200 14 27 47 57 58 58 61 56 57 57 57 61 D
jC2DSea2 D:SA 2400 7200 13 23 46 54 57 62 69 62 55 54 53 69 &D20 23 11 8 2 0 0
jC2DSea1 SA 2400 7200 5.9 11 18 25 35 49 55 59 62 72 71 72 D
iP4SeaRd SR 3633 15000 1.7 3.2 6.3 8.9 19 29 28 36 51 65 75 75 3 3 3 1 1 0 0
nTbdIBMRd IR 1400 7200 11 19 25 36 49 71 71 67 73 74 76 76 69 59 45 31 17 16 9
oPalSeaRd %SR 1700 15000 2.2 5.0 10 18 29 43 57 64 73 79 82 82 7 9 8 7 7 5 3
gP3IBMRd SR 1000 10000 1.3 2.4 4.1 6.0 8.9 15 30 65 81 91 44 91 D
iP4HiSRd AR 3000 7200 6.4 18 29 47 69 55 67 68 85 105 101 105 6 12 11 10 7 1 2
iP4WDSRd AR 3633 10000 14 24 37 54 68 65 89 103 108 109 108 109 52 43 34 23 15 6 5
jC2DSea3 &ES 2400 7200 11 20 39 60 81 98 115 115 112 112 115 115 &D30 30 28 27 19 13 8
at 145 GB &ES 2400 7200 11 20 36 58 79 97 99 98 97 98 96 99 &D28 28 24 21 15 11 5
at 290 GB &ES 2400 7200 11 20 36 58 66 67 64 65 63 65 63 67 &D32 27 24 21 14 7 5
qPheWD1 &SA 3000 5400 12 24 39 77 78 97 91 103 98 101 104 104 &Q13 12 14 14 11 9 7
at 224 GB &SA 3000 5400 16 28 49 84 92 95 94 94 93 93 91 95 &Q15 14 13 13 12 9 7
at 448 GB &SA 3000 5400 14 27 51 79 77 79 82 81 79 78 80 82 &Q12 13 12 10 8 5 5
q64ASeaR *SR 1991 15000 8.6 16 28 46 68 87 106 120 125 128 129 129 D
oPalMaxRd $IR 1667 7200 21 45 67 90 113 129 55 74 72 129 D
jI7ISSD SA 2800 SSD 25 45 74 98 120 131 158 192 198 206 217 217 Q
q64ASeaR2 *SR 1991 15000 6.2 12 23 41 60 95 97 173 193 209 221 221 D
% = PCI 50MHz SA = SATA SC = SCSI AR = SATA Raid IR = IDE RAID SR = SCSI Raid D = Dual CPU Q = Quad CPU
ES = eSATA $ = High Speed PCI * = PCI-X/64bit 133 MHz & = 64 bit version C:/D: 200 GB partitions
DMA Speed
CPU Disk MBytes/Second at KB Block Size % CPU Utilisation
Key MHz RPM 1 2 4 8 16 32 64 128 256 512 1024 Max 1 2 4 8 16 32 64
a486DX2 66 3600 1.2 2.1 3.3 0.5 1.0 0.7 1.0 1.1 1.2 1.3 1.3 3 100 100 100 18 42 27 43
dPProQua 200 4500 1.6 2.4 3.2 3.7 4.0 4.3 4.3 1.9 2.1 2.7 3.0 4 53 68 81 88 92 96 95
dPProQua 200 4500 1.6 2.4 3.3 3.6 4.1 4.3 4.4 2.3 2.7 3.1 3.6 4 54 67 80 85 93 96 98
eCelLap 400 4200 2.6 4.0 6.4 5.9 8.1 9.3 11 11 3.5 4.4 5.4 11 28 22 20 14 10 7 5
bPenSea 100 5400 4.2 7.3 11 2.7 2.1 1.9 1.8 1.6 1.6 1.7 1.8 11 100 100 100 98 99 100 99
K62Sea2 380 5400 1.7 3.9 5.3 8.5 11 14 5.6 5.6 5.6 6.4 6.9 14 29 38 23 20 18 12 3
fPIISea1 SC 450 10000 0.20 0.30 0.70 1.3 2.6 5.3 10 11 14 14 17 17 5 81 26 4 6 5 4
fPIIFuj 450 5400 2.3 4.2 7.7 12 17 17 20 3.8 5.6 7.5 7.8 20 24 22 21 19 11 9 5
K62Sea 500 7200 5.6 8.6 14 18 21 23 25 26 26 26 21 26 69 53 41 30 19 11 7
eCelMax 450 7200 3.0 8.4 11 17 22 25 26 23 10 12 12 26 28 39 25 19 12 6 5
gp3SeaA 1400 7200 6.8 11 15 20 23 25 26 27 26 20 16 27 32 26 17 13 6 3 2
gP3Max 1400 4.7 9.2 15 21 25 27 29 30 15 17 20 30 22 22 19 12 8 4 3
eCelMax2 700 5.1 9.0 14 19 24 27 29 30 30 19 20 30 43 36 26 18 9 5 3
eCelMax1a 950 7.4 12 17 23 26 28 29 30 30 22 30 30 53 43 31 18 9 7 2
cPMMXQua AV 200 5400 11 20 31 14 12 12 12 5.6 5.6 6.4 6.9 31 100 100 100 97 100 100 100
mDurMax2 W2K 950 7200 6.0 12 18 24 29 32 32 31 15 20 24 32 48 46 35 24 15 9 11
mDurMax1 W98 950 7200 3.3 6.7 13 20 27 32 36 34 15 20 24 36 24 24 24 20 12 7 4
oPalQuaS SC 1533 10000 0.20 0.30 0.60 1.3 2.6 5.2 10 21 20 28 39 39 D
eCelWD1 466 7200 8.4 14 24 34 43 50 28 47 15 30 40 50 89 81 64 46 33 21 9
iP4Sea2 1900 5400 6.0 13 21 31 39 45 51 50 11 15 18 51 29 30 24 18 10 5 1
oPalSea3 SC 1667 15000 5.7 10 18 27 37 45 51 52 52 41 41 52 D
iP4Sea2 N 1900 5400 5.7 13 20 28 37 45 49 52 11 52 30 52 34 32 25 20 10 5 3
oPalSea2 %SC 1700 7200 5.7 12 20 29 37 45 50 52 53 30 30 53 27 28 25 17 11 6 3
K62IBM 267 7200 5.9 11 20 29 39 48 53 51 51 57 57 57 100 100 100 90 68 48 36
oPalIBM3 1539 7200 11 22 34 43 50 55 58 57 58 58 58 58 73 74 56 38 20 11 5
iP4Hit1 1900 7200 11 21 34 44 51 56 59 61 61 61 61 61 58 64 42 27 14 7 2
nTbdIBM8 1400 7200 7.5 17 33 43 50 56 59 60 60 60 60 60 76 50 50 33 19 11 6
iP4Sea3 2533 7200 7.6 15 22 37 46 53 59 61 61 30 28 61 32 33 25 19 11 5 2
oPalMax1 1667 7200 4.6 8.6 16 27 41 53 62 68 71 70 28 71 D
nTbdMax1 1533 5400 6.4 13 22 34 48 60 69 72 11 23 22 72 16 15 13 10 7 4 2
iP4Sea1 2016 7200 7.5 15 27 40 52 63 70 74 70 30 34 74 55 57 51 37 24 14 8
iP4SeaRd SR 3633 15000 1.7 3.3 6.4 12 21 35 51 55 66 77 68 77 4 3 5 3 3 2 1
iP4Max 2400 5400 5.3 10 19 31 46 62 74 78 74 45 43 78 23 23 20 16 11 6 3
gP3IBM1 1000 7200 11 24 39 53 65 73 78 73 78 79 78 79 77 88 73 50 30 17 9
nTbdQuaRd IR 1400 7200 3.9 7.2 13 24 38 54 63 73 78 80 79 80 65 62 58 52 42 30 44
oPalSeaR0% SC 1700 15000 2.3 4.6 8.7 16 27 42 57 71 77 80 52 80 7 7 7 6 5 4 2
iP4NLap 1700 5400 4.9 10 17 30 42 57 70 80 83 82 75 83 34 35 30 26 17 11 7
nTbdQua 1200 7200 6.2 11 21 35 53 65 74 77 81 84 83 84 26 23 22 19 14 10 5
iP4Sam 3066 5400 6.7 13 23 36 53 67 78 85 83 60 26 85 18 17 15 13 8 6 3
pBarLap 1800 4200 6.7 14 19 34 52 66 77 83 85 86 80 86 24 25 17 15 11 7 4
iP4Sea4 2533 7200 8.5 24 32 49 65 78 83 88 87 23 22 88 38 53 35 26 16 9 3
iP4WD1 2411 7200 13 27 40 57 70 79 32 87 88 30 60 88 48 50 37 26 16 9 0
ip4WD2 2411 7200 15 27 40 57 70 79 84 87 88 30 40 88 57 50 37 26 16 8 4
iP4WD3 2533 7200 14 19 39 55 69 79 85 89 89 39 40 89 66 82 45 31 18 17 4
oPalWD1 1410 7200 12 21 32 50 65 77 83 89 89 89 40 89 71 66 100 69 28 15 15
q64ASea1 2150 7200 8.8 18 32 49 65 77 86 91 89 30 40 91 23 23 20 15 10 5 2
oPalWD2 1466 7200 15 25 37 56 70 80 87 91 91 91 48 91 59 50 50 28 18 11 5
oThoWD3 2300 7200 15 28 44 59 73 30 89 92 92 76 40 92 51 49 38 27 16 3 5
iP4WD4 3000 7200 13 32 49 65 77 85 90 55 92 92 40 92 31 37 33 18 10 4 1
oPalIBM2 1466 7200 16 30 49 65 77 85 90 92 92 92 91 92 56 53 42 28 16 9 5
oThoHi1 2080 7200 12 25 48 71 78 87 91 93 93 93 93 93 28 31 29 22 12 7 3
gP3Sea15 $SC 667 15000 5.6 11 21 30 57 74 99 109 111 45 49 111 D
jC2DLap SA 1830 5400 5.3 15 16 43 64 81 97 107 110 113 112 113 D
oPalIBMRd $IR 1396 7200 13 26 42 58 72 89 111 112 112 113 113 113 82 92 75 52 32 35 17
jC2DSea3 D& ES 2400 7200 11 20 39 61 82 93 99 118 109 107 108 118 35 31 31 24 21 11 13
iP4EWD1 SA 3000 7200 14 25 42 57 80 99 112 117 119 119 117 119 7 2 2 13 1 2 2
jC2CLap SA 2000 5400 5.6 12 27 36 66 89 108 120 121 118 118 120 32 33 36 23 27 16 8
q64TLap SA 1900 5400 5.7 17 31 50 73 95 112 123 124 125 115 125 D
q64AMax1 &SA 2210 7200 23 39 62 86 105 119 127 131 38 59 60 131 D16 20 18 19 8 7 3
q64AMax1 SA 2210 7200 22 39 60 85 104 118 125 131 41 29 38 131 D
q64ASeaR *SR 1991 15000 5.3 11 19 33 54 79 103 119 115 130 131 131 D
oPalMaxRd $IR 1667 7200 7.9 15 27 44 65 93 114 114 145 153 158 158 D
jI7ISSD SA 2800 SSD 8.0 11 16 36 54 89 128 169 172 173 164 173 Q
iP4WDSRd AR 3633 10000 16 28 45 62 32 88 95 135 188 42 83 188 60 55 41 28 5 9 4
qPheWD1 & SA2 3000 5400 9.3 19 33 60 88 130 166 186 197 203 204 204 Q 9 10 7 10 11 10 7
iP4HiSRd AR 3000 7200 9.8 20 42 71 91 105 126 183 204 207 208 208 19 20 20 17 11 6 7
jC2DSea2 & SA2 2400 7200 13 26 49 80 120 165 205 220 220 219 206 220 D31 29 31 28 17 10 12
jC2DSea2 SA2 2400 7200 12 25 50 82 133 165 206 235 236 227 209 236 D
q64ASeaR2 *SR 1991 15000 6 12 23 43 74 117 158 196 209 145 233 233 D
jC2DSea1 SA2 2400 7200 15 18 35 68 109 145 173 208 229 237 243 243 D
% = PCI SA = SATA SA2 = SATA2 SC = SCSI AR = SATA Raid IR = IDE RAID SR = SCSI Raid D/Q = Dual/Quad CPU
ES = eSATA * = PCI-X/64bit 133 MHz $ = High Speed PCI AV = Norton AV in use & = 64 bit version
|
To Start Results/Examples Index
Random Access Time Small Files
Buffer CPU Disk Milliseconds at MB file size
Key MB MHz RPM 1 2 4 8 16 32 64
a486DX2 66 3600 12.7 19.1 21.4
dPProQua S 200 4500 9.9 11.0 10.6 16.0
dPProQua S 200 4500 10.0 12.9 12.5 12.8
eCelLap 400 4200 9.9 9.7 9.6 11.9
K62Sea2 E 380 5400 8.5 7.6 9.3
cPMMXQua S 200 5400 4.9 7.4 8.2 8.6
fPIIFuj Q 450 5400 7.7 7.9 8.1 8.5
pBarLap 2 1800 4200 1.3 4.5 6.3 8.3
oPalWD1 1410 7200 0.5 1.9 6.7 7.1
eCelNA2 900 5.5 5.7 5.9 7.1
iP4Sam 3066 5400 4.9 6.6 6.1 7.0
iP4Sea2 2 1900 5400 4.2 5.8 6.0 7.0
eCelMax1a 950 1.2 3.4 5.4 6.9
nTbdMax1 2 1533 5400 2.0 4.4 6.3 6.8 7.5
eCelMax Q 450 7200 6.5 6.9 6.2 6.5
iP4Sea2 2 1900 5400 1.9 5.9 6.5 6.4
gP3Max 1400 1.9 3.9 5.0 6.0
jC2CLap SA 2000 5400 0.6 0.3 2.0 5.9 8.5
iP4Max 2400 5400 2.2 3.9 5.8 5.8
iP4Sea1 2 2016 7200 0.9 3.1 4.2 5.7
gp3SeaA 1400 7200 1.3 2.7 5.1 5.6
nTbdQuaRd 2R? 1400 7200 0.6 1.7 4.4 5.6
iP4Sea3 2 2533 7200 1.0 3.0 4.2 5.5
nTbdIBM8 2 1400 7200 0.7 2.2 4.1 5.1
K62Sea 2 500 7200 0.6 3.3 4.3 5.0
oPalSea2 2 1700 7200 0.7 3.0 3.5 4.9
iP4Sea4 8 2533 7200 0.4 1.0 2.7 4.9
mDurMax2 2 950 7200 0.8 2.6 4.3 4.9
mDurMax1 2 950 7200 0.8 2.3 4.2 4.8
nTbdQua 2 1200 7200 0.9 3.0 3.8 4.8
eCelWD1 466 7200 0.4 2.0 4.2 4.7
gP3IBM1 2 1000 7200 0.7 2.7 3.5 4.7 5.4
iP4EWD1 8 3000 7200 0.2 0.9 2.4 4.5
oPalMax1 1667 7200 2.1 3.5 4.4 4.4
K62IBM 2 267 7200 0.6 1.7 4.2 4.4
iP4Hit1 2 1900 7200 0.7 2.6 3.8 4.4
oThoHi1 2 2080 7200 0.3 2.3 3.5 4.4
oPalIBM3 2 1539 7200 0.5 2.7 3.8 4.3
fPIISea1 1 450 10000 6.0 5.0 4.8 4.3 4.7
oPalIBM2 2 1466 7200 0.4 2.7 3.7 4.1
q64ASea1 8 2150 7200 0.4 0.8 2.0 3.9
oPalIBMRd 2R2 1396 7200 0.1 0.6 1.9 3.8
eCelMax2 700 1.9 2.8 1.8 3.8
iP4NLap 16 1700 5400 0.2 1.0 2.8 3.3
q64TLap 8 1900 5400 0.2 0.2 1.1 3.2
jC2DSea2 16 2400 7200 0.1 0.5 1.2 3.0 6.1 5.2 5.4
oPalSeaR0 8 1700 15000 0.4 1.1 2.2 2.9
oThoWD1 8 1737 7200 0.1 0.2 0.7 2.8
iP4WD4 8 3000 7200 0.1 0.2 0.8 2.8
jC2DSea1 16 2400 7200 0.2 0.1 0.5 2.8
iP4WD3 8 2533 7200 0.1 0.2 0.6 2.6
oPalSea3 1667 15000 0.2 0.7 2.1 2.4
oPalWD2 8 1466 7200 0.1 0.1 0.5 2.3
iP4SeaRd 8 3633 15000 0.8 0.7 0.9 2.1
gP3Sea15 8 667 15000 0.3 0.6 1.3 2.0
q64ASeaR 3S8 1991 15000 0.2 0.4 0.6 1.5
ip4WD2 8 2411 7200 0.1 0.2 0.8 1.2
iP4WD1 8 2411 7200 0.1 0.1 0.6 1.1
iP4HiSRd 2R8 3000 7200 0.1 0.3 0.7 0.9
q64AMax1 16 2210 7200 0.04 0.04 0.2 0.9 2.9 4.1 4.9
q64ASeaR2 4S8 1991 15000 0.2 0.3 0.5 0.9 1.3 1.8
iP4WDSRd 2R8 3633 10000 0.1 0.1 0.3 0.9 1.8 3.4
jC2DSea3 & 16 2400 7200 0.3 0.4 0.4 0.4 2.8 4.8 6.1
q64AMax1 & 16 2210 7200 0.04 0.04 0.04 0.05 0.7 5.0 6.4
2R = 2 Disk SATA Raid 3S/4S = 3/4 Disk SCSI Raid SA SATA
S = Small Q = 256 KB E = 128 KB & = 64 bit version
|
|
To Start Results/Examples Index
Extended Random Access Times
Following are random reading times of files up to 1 GB. The laptop disk runs at 5400 RPM and, except for the WD Green drive, the others run at 7200 RPM. The results are influenced by buffer size, where speeds are very fast when data is cached there, but with reducing impact as the file size increases. Then latency (half disk revolution time) come into play at 4.2 milliseconds for 7200 RPM and 5.6 milliseconds at 5400 RPM. Finally, head switching and movement affects larger files, influenced by the number of heads/platters and amount of data per track. Fragmentation will also have some impact.
Results for a SSD are also included, showing a constant 0.1 milliseconds random reading time.
The WD green disk (qPheWD1) performs well on these tests, initially casting some doubt that the rotation speed was 5400 RPM. Then tests were run on the tiny WD Passport USB drive (yDiskUSB2g) where performance is similar on the larger files.
Buffer CPU Disk Milliseconds at MB file size
Key MB MHz RPM 1 2 4 8 16 32 64 128 256 512 1024
F: yDiskUSB2g 8 2400 5400 0.4 0.4 0.4 2.8 5.2 6.4 7.5 7.5 8.1 7.9 8.2
C: jC2DLap 8 1830 5400 0.2 1.5 2.4 5.4 9.1 9.5 9.5 9.8 10.8 9.6 10.5
D: jC2DLap 8 1830 5400 0.3 1.5 2.6 6.0 6.4 7.2 8.4 8.2 8.6 9.6 10.6
C: qPheWD1 & 16 3000 5400 0.1 0.1 0.1 0.1 0.3 3.7 6.3 6.9 7.2 7.0 8.4
D: qPheWD1 & 16 3000 5400 0.1 0.1 0.1 0.1 0.1 2.8 5.4 7.4 7.6 7.7 7.6
E: qPheWD1 & 16 3000 5400 0.1 0.1 0.1 0.9 2.4 4.5 6.0 8.2 8.1 8.1 8.0
C: q64AMax1 16 2210 7200 0.04 0.04 0.04 0.05 1.3 5.2 7.4 8.7 9.1 9.2 9.6
C: jC2DSea2 & 16 2400 7200 0.1 0.1 0.3 2.5 6.3 8.0 8.1 7.3 6.1 6.3 6.7
D: jC2DSea2 & 16 2400 7200 0.1 0.2 0.2 3.6 5.1 5.7 5.9 6.0 5.7 6.0 6.7
G: jC2DSea3 & 16 2400 7200 0.3 0.4 0.4 0.5 2.7 5.1 5.8 6.3 6.9 7.9 8.1
H: jC2DSea3 & 16 2400 7200 0.3 0.4 0.4 0.5 2.9 5.4 6.2 7.0 6.9 7.0 7.2
I: jC2DSea3 & 16 2400 7200 0.3 0.4 0.4 0.8 3.4 5.3 6.4 6.6 7.3 7.4 7.5
C: jI7ISSD 2800 SSD 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1
Partitions GB jC2DLap C: 70, D: 70, qPheWD1 C: 224, D: 224, E: 146,
jC2DSea2 C: 195, D: 177, jC2DSea3 G: 145, H: 145 I: 30 eSATA
yDiskUSB2g C: 250
|
To Start Results/Examples Index
|
Configuration Key
Code Configuration
a486DX2 80486DX2 66.7 MHz, 3600 RPM, Win95 (1994)
bPenSea Pentium 100 MHz, Seagate ST5850A, 850MB, 3600 RPM Disk, Win95, ISA connection (1995)
cPMMXQua Pentium MMX 200 MHz, Quantum Fireball ST 2.1AT, 2.1GB, 5400 RPM disk, Win95 (1997)
dPProQua Pentium Pro 200 MHz, Quantum Fireball TM2100A, 2.1GB, 4500 RPM disk, NT 4.0 (1995)
dPProQua Second result version 1.3A
eCelLap Celeron 400 MHz Laptop, Fujitsu MFH 2043, 4.3 GB 4200 RPM disk, Win98 (1999)
eCelMax Celeron 450/100MHz, Maxtor DiamondMax Plus 2500 7200 RPM disk, Win98 (1999)
eCelMax1 Celeron 950/100 MHz, Win98, Maxtor DMA66 disk, 5400 RPM (2001)
eCelMax1a As CelMax1, DMA not working WinXP, different Maxtor disk (2003)
eCelMax1a Second result DMA working via reinstalling primary IDE
eCelMax2 Celeron 700 MHz, 80 GB Maxtor disk, UDMA 33. Results via Win98 and WinXP (2003)
eCelNA2 Celeron 900/100 MHz, WinME (2002)
eCelWD1 Celeron 466 MHz, WD 60 GB 7200 RPM disk, win98 (2002)
fPIILap Laptop, Mobile Pentium II/PE 300 MHz, 5400 RPM disk, Win98 (1999)
fPIIFuj Pentium II 450 MHz, 512K cache, Win98, Fujitsu MPC 3102AT 10.2GB 5400 RPM Disk (1999)
fPIISea1 PII 450, Win98, LVD 10Krpm Seagate ST39102LW SCSI, 9GB (1998)
gP3IBM1 PIII 1000 MHz, IBM Deskstar 60GXP, 40GB, 2MB cache, 7200 rpm, ATA100: DMA not enabled (2001)
gP3IBM1 DMA enabled
gP3IBMRd Dual PIII 1 GHz, RAID0, 4 x IBM 10K RPM SCSI160 18.3 GB, Adaptec 3200 64 bit RAID card 80 MB
cache, Win2K (2001)
gP3Max Powerleap CPU upgrade to 1400 MHz, 20GB Maxtor disk, Win2K (2003)
gP3RAID Dual Pentium III 1 GHz, IDE RAID0, Win2K (2002)
gP3Sea15 Dual Xeon 667 MHz, Seagate Cheetah Ultra 160 SCSI 15K, ST318452LW, Adaptec AIC-7899,
Win2K (2002)
gp3SeaA Powerleap CPU upgrade 1400 MHz, 40GB Seagate Barracuda, 7200 RPM, ATA/100 drive mobo
ATA/33 Win2K (2003)
gP3WD1 PIII 1000, Intel drivers, WD400BB, 7200 RPM, partitions C, D, E (2001)
gPIIILap Toshiba Satellite 2800 Laptop, PIII 650 MHz, WinME (2000)
iP4EWD1 P4E Prescott 3.0 GHz, WD2000JD 7200 RMP, 8 MB buffer, 200 GB SATA disk (3/4 full), WinXP (2004)
iP4Hit1 P4 1900 MHz, Hitachi 7200 RPM 7K80, WinXP
iP4HiSRd P4 3 GHz, two 80GB Hitachi 7K250 Deskstars, 8MB cache, SATA RAID 0, 128K stripe on
Intel ICH5R WinXP (2004)
iP4Max P4 2405 MHz, Maxtor 5400 RPM, 80 GB drive (D:) Normal results + same file cache enabled (2003)
iP4NLap Laptop, 1700 MHz P4-M Northwood CPU, WinXP, Toshiba MK4019GAX 20 GB disk 5400 RPM (2003)
iP4Sam P4 3.06 GHz, Samsung SV08134 disk, 80 GB, 5400 RPM, Win XP (2003)
iP4Sea1 P4 2016 MHz, Seagate Barracuda IV 7200 RPM disk, WinXP (2001)
iP4Sea2 P4 1900 MHz, 3 partitions, Seagate ST340810A 40GB 5400 RPM disk, WinXP (2001)
iP4Sea2 2 other partitions and using Win File Cache
iP4Sea3 P4 2.53 GHz, Seagate ST380021A Barracuda 4 80 GB 7200 RPM, WinXP (2003)
iP4Sea4 P4 2.53 GHz, Seagate Barracuda V, ST3120024A, 8MB cache, WinXP (2003)
iP4SeaRd P4 3633 MHz, 2 Seagate Cheetah 15K RPM disks, Adaptec cotroller, RAID0, WinXP (2003)
iP4WD1 P4 2411 MHz, Western Digital 120GB disk 8MB cache, 7200 RPM, WinXP (2002)
ip4WD2 P4 2411 MHz, WD 100GB 8MB Intel and MS drivers, 7200 RPM, WinXP (2002)
iP4WD3 P4 2.53 GHz, Western Digital 80GB HD 8MB cache, WinXP (2002)
iP4WD4 P4 3.0 GHz, WDC WD800JB 7200 RPM 8 MB buffer, Win2K (2003)
iP4WDSRd P4 3363 MHz Abit 1C7G Canterwood,SATA RAID0, 2 WD Raptor disks, 10,000 RPM/8MB cache (2003)
iP4Xeon Dual Xeon 2.2 GHz, IDE disk, Win2K (2002)
jC2CLap Compaq 6720s, Celeron 550 2.0 GHz, Fujitsu 5400 RMP Disk, Vista Home Basic (2008)
jC2DSea1 Core 2 Duo 2.4 GHz, Seagate ST3320620AS SATA Disk 320 GB, 16 MB buffer, WinXP (2006)
jC2DSea2 Core 2 Duo 2.4 GHz, Seagate ST3400633AS SATA Disk 400 GB, 16 MB, Vista 64 Bit (2007)
4 GB DDR2 RAM, 32 bit and 64 bit results
jC2DSea3 As jC2DSea2 but Seagate ST3320613AS 320 GB, in Akasa 1.5 Gb/sec eSATA enclosure (2009)
Results early tracks, at 145 GB and 300 GB
jC2DLap Acer 7720G, Core 2 Duo 1.83 Ghz, Hitachi HTS5425 5400 RPM disk, 8 MB cache, Vista 32 Bit (2008)
jI7ISSD Intel Corei7 860 2.8Ghz, Intel X25-M 80GB SSD, 64-Bit Windows 7 (2009)
K62IBM AMD K6-2/266MHz, IBM Deskstar 80GB 7200rpm 2MB cache, Win98 (2002)
K62Sea AMD K62 500 MHz, Seagate Barracuda 20GB drive, Win2K (2001)
K62Sea2 AMD K6-2 380 MHz, Seagate ST36531A 6.5GB, Win98
K63WD AMD K6-3 400 MHz, WD136AA 5400 RPM disk 13 GB, Win 98SE (2000), slow with DMA on
K63WD Second results DMA off and third DMA on with new IDE driver (2000)
mDurMax1 Duron 950 MHz, Win98SE, Maxtor DiamondMax Plus 40 53078H6, 7200 RPM, IDE UDMA66 (2000)
mDurMax1 Second results using Win File Cache (2000)
mDurMax2 As DurMax1, but Win2K (2000)
mDurMax2 Second results using Win File Cache (2000)
nTbdIBM8 Tbird 1.4 GHz, IBM 180GXP 60GB 2 MB buffer 7200 RPM, Win2K (2003)
nTbdIBMRd Tbird 1.4 GHz, Promise FastTrak 100-TX2 EIDE RAID, 2 x IBM 60GXP, 7200 RPM, 2MB, Win2K, (2001)
nTbdMax1 Tbird 1544 MHz, Maxtor 60GB, ATA100, 2MB, 8ns, 5400RPM, NT4 (2001)
nTbdMax1 Second results, drivers re-installed - faster writing (2001)
nTbdQua AMD Tbird 1200 MHz, Quantum Fireball AS 20 GB 7200 RPM ATA100, Win98 (2001)
nTbdQuaRd Athlon 1.4 GHz RAID0 with the KK266R, 2-40Gb Quantum Fireball 7200 rpm, WinXP (2002)
nTbdWDI1 Tbird 1296, VIA driver, WD400BB 7200 RPM, partitions C, D, E (2001)
oPalIBM2 AMD XP 1700 1.47 MHz, IBM disk 60GB 7200 rpm 2mb buffer,. Win2K (2001)
oPalIBM3 AMD XP1800 1539 MHz, IBM Deskstar 120GXP 41.2GB UDMA 100, Win98 (2003)
oPalIBM3 Second result NAV enabled, third result IDE socket instead of RAID
oPalIBMRd Athlon 4 1397 MHz, partitions C, D, E, 2-IBM 60GXP 40GB, RAID 0, HPT 370A, Win2K (2001)
oPalMax1 Dual athlon 1.67 Ghz, Maxtor ATA133 80GB 7200, onboard primary IDE, Win2K (2002)
oPalMaxRd Dual athlon 1.67 Ghz, 3ware: Escalade 7810 w/ 8 Maxtor ATA133 160GB 5400, raid 0, Win2K (2002)
oPalQuaS Dual AMD MP 1533 MHz, Quantum 10K RPM disk, Adaptec 29160 U160, WinXP, (2002)
oPalSea2 Athlon XP 1700 MHz, Barracuda IV, 7200 RPM, Adaptec 2110S, SCSI Ultra 160, WinXP (2002)
oPalSea3 Dual Athlon 1667 MHz, Adaptec 19160 SCSI, 15K RPM Seagate Cheetah disks, Win2K (2002)
oPalSeaR0 As oPalSeaRd PCI but single Seagate 15K RPM Cheetah ST318452LC, PCI 50 MHz (2002)
oPalSeaR2 As oPalSeaRd PCI but bus at 34 and 42.5 MHz (2002)
oPalSeaRd Athlon XP 1700 MHz, RAID0 Adaptec 2110S, SCSI Ultra 160, 2 x Seagate 15K RPM ST318452LC,
18.4 GB, 8 MB buffer, PCI 50 MHz, WinXP (2002)
oPalWD1 Athlon XP 1410 MHz, ASUS A7V266-E, WD40G-7200 RPM (2002)
oPalWD2 AMD XP 1.47 GHz, SLOW, Western Digital WD1200JB-00CRA1, Win98 (2003)
oPalWD2 Fast results WinXP
oThoHi1 AMD Tbred 2.08 GHz, Hitachi 7K250 123 GB, 7200 RPM, 2 MB, Win2K (2004)
oThoHi1 More results using nVidia IDE driver instead of MS
oThoHi2 AMD Tbred 2.08 GHz, Hitachi 7K250 123 GB, 7200 RPM, 2 MB, Win98 (2004)
oThoHi3 AMD Tbred 2.38 GHz, Hitachi 7K250 123 GB, 7200 RPM, 2 MB, Win2K (2004)
oThoHi3 More results using nVidia IDE driver instead of MS
oThoWD1 AMD Tbred 1.73 GHz, WD 120 GB disk, 7200 RPM, 8 MB cache, WinXP (2003)
oThoWD3 AMD Tbred 2.30 GHz, WD 120 GB disk, 7200 RPM, 8 MB cache, WinXP (2003)
pBarLap Laptop, AMD XP 1800 MHz, WinXP, Toshiba MK4018GAS 40 GB disk, 4200 RPM, 2MB (2003)
q64AMax1 Athlon 64 X2 2.21 GHz, 300 GB Maxtor disk 7200 RPM SATA 16 MB Buffer, WinXP Pro X64 (2005)
q64ASea1 Athlon 64 2.15 GHz, 120GB Seagate Barracuda V, 8MB cache, Win XP (2003)
q64ASea2 As q64ASea1 with MS driver plus new IDE 2.0 nVidia driver 3.66
q64ASeaR Dual Opteron 246 CPUs 2 GHz, 3 x Seagate 15K RPM Cheetah ST318452LC, 18.4 GB, 8 MB buffer,
Adaptec 39320D-R PCI-x/64bit 133 Mhz, WinXP (2003)
q64ASeaR2 Dual AMD Opteron 246 CPUs 2 GHz, 4 x Seagate 15K RPM Cheetah ST318453LC ultra 320 disks,
18.4 GB, 8 MB buffer, Adaptec 39320D-R PCI-x/64bit 133 Mhz, WinXP (2003)
q64TLap Acer 5520G Laptop, Turion 64 X2 1900 MHz, WD2500BEVS disk 5400 RPM, Vista (2007)
qPheWD1 Phenom II X4 3.0 GHz, 640 GB WD 5400 RPM WD6400AACS disk, 16 MB buffer, 64-Bit Windows 7 (2009)
yDiskUSB2a P4 1900 MHz, Freecom Disk 7200 RPM, 8 MB buffer, (Hitachi HDT72252 5DLAT80), WinXP
yDiskUSB2b Athlon 64X2 2.21 GHz, Freecom Disk 7200 RPM, 8 MB buffer, (as above) WinXP x64
yDiskUSB2c AMD Tbred 2.08 GHz, Freecom Disk 7200 RPM, 8 MB buffer, (as above), Win2K
yDiskUSB2f Core 2 Duo 2.4 GHz, Freecom Disk 7200 RPM, 8 MB buffer, (as above), 64-Bit Vista
yDiskUSB2g Core 2 Duo 2.4 GHz, WD 2500MER Passport 250 GB, 5400 RPM, 64-Bit Vista
yDiskUSB2d External USB 2.0 Acomdata box with WD disk, AMD CPU, Win2K.
yDiskUSB2e Same disk USB 2.0 and Firewire, P4 3 GHz
yDiskFwire Same disk USB 2.0 and Firewire, P4 3 GHz
zCDRW Duron 950 MHz, Iomega ZipCD USB Drive (6 x 4 x 4) via benchmark on CDRW disc, Win98
zCFlashLP Duron 950 MHz, San Disk Compact Flash Card Reader on LP port. Benchmark on 20 MB CF, Win2K
zCFPcmcia Celeron 400 MHz Laptop, Compact Flash in PCMCIA adapter, Win98
zCFPcmci2 AMD XP 1.8 GHz Laptop, Compact Flash in PCMCIA adapter, WinXP
zDVDRW Pentium 4 1.9 GHz, Liteon LH-20A1H DVD +RW 8x, disc 4x, Nero InCD 5, WinXP
zFD4096A Staples USB Flash Drive 4 GB
zFD4096C SanDisk Cruzer Micro 4 GB, Enhanced for ReadyBoost
zLAN10mbs Disk on remote Duron 950 PC. Test on Celeron 450 MHz via 10 Mb/sec Lan, Win98.
zLAN100m1 100mbs LAN disk on remote Duron/Win2K server, program run on P4 1900 WinXP
zLAN100m2 100mbs LAN disk on remote P4/WinXP server, program run on Duron/Win2K
zLAN100m3 100mbs LAN disk on remote Fujitsu Laptop WinXP, program run on P4 1900 WinXP
zSD2GB System jC2DSea2, 2 GB SanDisk Ultra II 110x speed rating SD Card, SanDisk Reader (2009)
zSD32MB System jC2DSea2, 32 MB SD Card supplied with Canon camera, SanDisk Reader (2009)
zUSB1CF P4 1.9 GHz, WinXP, Acrox Compact Flash Card Reader, old card, USB 1.1
zUSB1CF P4 1.9 GHz, WinXP, Acrox Compact Flash Card Reader, new card, USB 1.1, 2003
zUSB2CF P4 1.9 GHz, WinXP, Acrox Compact Flash Card Reader, old card, High Speed USB 2
zUSB2CF P4 1.9 GHz, WinXP, Acrox Compact Flash Card Reader, new card, High Speed USB 2, 2003
zUSBComms Disk on remote Duron 950 PC. Test run on P4 1.9 GHz WinXP. Link Belkin USB Direct Connect
zWiFi56 jC2DLap above wireless connection 56 kbps
|
|
|
