At 10:27 PM -0500 8/21/2010, Kris Tilford wrote:
On Aug 21, 2010, at 9:32 PM, Dan wrote:
SeriTek 1V4 is SATA I, 1.5 Gbps burst, 150 MB/sec (1.2 Gbps)
nominal on a good day.
This isn't exactly right. Copied below [snip] SeriTek/1VE2+2
The card in question is a **** 1V4 **** not a 1VE2+2.
The specs were obtained from FirmTek's spec page pertaining to the 1V4.
At 8:00 PM -0700 8/21/2010, John Carmonne wrote:
On Aug 21, 2010, at 7:32 PM, Dan wrote:
At 2:58 PM -0700 8/21/2010, John Carmonne wrote:
SeriTek 1V4 in my PM G4 and PM G5 the transfer speed is 88 sec's
per GB on a 114 GB CCC. I was under the impression it would a lot
faster than this I seem to remember a 3 GB per min claim.
Both HDD's are 2TB 7200 RPM Hitahci's in the PM G5.
SeriTek 1V4 is SATA I, 1.5 Gbps burst, 150 MB/sec (1.2 Gbps)
nominal on a good day.
eSATA, by its nature, runs a bit slower. What model Hitachi drives?
Exactly what type of backup were you doing? File or block
oriented? Virgin or a merge into an existing volume?
Virgin copy with CCC
Again, file or block oriented - "Incremental" or "Backup everything"?
This makes a big difference. The file oriented backup is moving one
file at a time and building a *new* file system, whereas the
everything mode is moving whole streams of blocks and doing no file
system work (it just copies the file system itself as a stream of
blocks too - including whatever errors be there). The latter should
be much faster.
114 GB DVD images and toast images is taking 2.8 hours
Not sure what that means. Are you saying 114 GB total, files that
are DVD images, or are you saying 114 files, 4.2 to 4.7 GB each or
are these double-layer (8 GB) images or ? How much actual data are
you talking about?
I think USB will do that but I haven't tried that to be sure. The
enclosure for the eSATA drive is an inexpensive SABERENT USB eSATA
IDE & SATA model could that be a problem? The machine is a PM G5
Dual 2.7. The Drives are as follows
[snip newegg url]
Deskstar 7K2000, 2 TB, 7200 rpm, 32 MB cache. 3 Gbps SATA III
interface. The media is rated at 1621 Mbps max (202 MB/sec). That
"max" is basically a mixed mode (r and w) burst rate. It benchmarks
at around 100 to 160 MB/sec write. ...Remember that writing is
always slower than reading, because it has to load the cache then
perform the actual write, whereas reading can take advantage of
pre-loading etc.
Ok, so figure a nominal write of 120 MB/sec. That's about 1 GB every
8 1/2 secs.
Note of course that CCC's throughput is NOT a good measure of a
drive. It's doing a lot of file system work and error checking.
Also make sure that Spotlight's indexing is disabled on the target -
otherwise it will be massively slowing down CCC, trying indexing
things as CCC writes.
As for your external boxes throughput,,, no telling there without
some serious testing. Some subjective observations: I've noticed my
cheap external boxes run at about half to 3/4 the speed on USB as my
more-expensive LaCie d2 Quadra box. Firewire speeds aren't as bad
but still slower. Haven't tried eSATA.
Back to your OP... "88 sec's per GB" is a rather odd way of saying
throughput. Where did you get that? With nothing else running on
your system, try observing the throughput using Activity Monitor's
Disk Activity pane...
A better way to check the throughput of a drive set-up, I think,
would be to use a dd command from Terminal. It can be used to test a
single drive by filling a file with zeroes, then it automatically
reports the throughput.
First, create a fairly large test file and put it on the device you
want to test.
dd if=/dev/zero bs=1024 count=1048576 of=/Volumes/MyTestHD/gigabyte.file
When finished, dd will report its performance. Since the data (all
zeros!) came from the aether, there's no "read" time - this is pure
write.
Now read that 1 GB file and throw the bits back into the aether, so
you have pure read timing.
dd if=/Volumes/MyTestHD/gigabyte.file of=/dev/null
( "if" stands for input file, "of" is output file)
Note that it will take a little while for these commands to run,
seeing as they're flinging around a billion bytes. If you hit ^T
(control t), some intermediate status info will be shown.
You could use any large file to do the test. The first dd command
above is just an easy way to create a big one.
Of course, the test results will be more accurate if your i/o buses
are NOT busy doing other things. So be sure to quit your apps...
HTH,
- Dan.
--
- Psychoceramic Emeritus; South Jersey, USA, Earth.
--
You received this message because you are a member of G-Group, a group for
those using G3, G4, and G5 desktop Macs - with a particular focus on Power Macs.
The list FAQ is at http://lowendmac.com/lists/g-list.shtml and our netiquette
guide is at http://www.lowendmac.com/lists/netiquette.shtml
To post to this group, send email to g3-5-list@googlegroups.com
For more options, visit this group at http://groups.google.com/group/g3-5-list
