|
Home > Archive > alt.certification.a-plus > December 2003 > 7200 RPM HD
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
|
|
| PLAIN JOE 2003-12-21, 11:24 am |
| I have a 4-year-old PC-100 with a 5400 RPM hard-disk computer and all the
new HDs are 7200 RPM.
Can I run my PC-100 computer with this faster hard-disk?
| |
|
| PLAIN JOE wrote:
> I have a 4-year-old PC-100 with a 5400 RPM hard-disk computer and all
> the new HDs are 7200 RPM.
> Can I run my PC-100 computer with this faster hard-disk?
yes
| |
| PLAIN JOE 2003-12-25, 1:24 pm |
| "Geoff" <fooooooool@hotmail.com> wrote in message
news:bs4krf$984bu$1@ID-17404.news.uni-berlin.de...
> PLAIN JOE wrote:
>
> yes
>
The other way, can I run a PC 133 computer with the slower 5400 RPM disk?
| |
|
| Apples and oranges. The spindle speed of your hard drive and your memory
clock speed have no effect on each other. The drives should work fine on
either computer as long as they are compatible with your IDE/ATA interface.
| |
| PLAIN JOE 2003-12-25, 6:23 pm |
| "JBS" <noone@thisdontwork.net> wrote in message
news:01GGb.8883$IM3.8466@newsread3.news.atl.earthlink.net...
> Apples and oranges. The spindle speed of your hard drive and your memory
> clock speed have no effect on each other. The drives should work fine on
> either computer as long as they are compatible with your IDE/ATA
interface.
>
What will be difference in a newer system, PC 133, running on a 5400 RPM
disk or 7200 RPM?
| |
| PLAIN JOE 2003-12-25, 6:23 pm |
| ())
"PLAIN JOE" <not@here.com> wrote in message
news:BoJGb.19708$Pg1.12934@newsread1.news.pas.earthlink.net...
> "JBS" <noone@thisdontwork.net> wrote in message
> news:01GGb.8883$IM3.8466@newsread3.news.atl.earthlink.net...
memory[color=blue]
on[color=blue]
> interface.
> What will be difference in a newer system, PC 133, running on a 5400 RPM
> disk or 7200 RPM?
>
Any noticeable speed differences?
| |
| SBFan2000 2003-12-25, 9:24 pm |
| obviously the data can be accessed faster when the platters are moving
faster! Would you notice it?? Probably not but, whether you notice or not,
it would be quicker and with the very little difference I would say, get the
7200RPM
"PLAIN JOE" <not@here.com> wrote in message
news:ZqJGb.19712$Pg1.2203@newsread1.news.pas.earthlink.net...
> ())
> "PLAIN JOE" <not@here.com> wrote in message
> news:BoJGb.19708$Pg1.12934@newsread1.news.pas.earthlink.net...
> memory
> on
RPM[color=blue]
> Any noticeable speed differences?
>
>
| |
| Barry Watzman 2003-12-25, 9:24 pm |
| The speed of a drive is transparent to a computer, as is the data rate
(e.g. ATA-66/100/133). Everything works with everything. Obviously,
faster is faster, but as far as working, there are just no issues.
PLAIN JOE wrote:
> "Geoff" <fooooooool@hotmail.com> wrote in message
> news:bs4krf$984bu$1@ID-17404.news.uni-berlin.de...
>
>
> The other way, can I run a PC 133 computer with the slower 5400 RPM disk?
>
>
| |
|
| Typically the 7200 RPM drives have faster access times and higher overall
throughput. However, there are a number of factors to consider (interface,
buffer size, PIO modes, etc) and higher spindle speed doesn't always
translate into significantly better performance. If you have a 5400 RPM
drive that works well, use it. I still have a couple of these around the
house that just keep going. If you are buying a new drive, you may want to
consider what will work best in your next system...then go for a 7200 or
even 10000+ RPM drive. You may need a separate SCSI or SATA controller card
to run the last.
You may notice a difference between the 5400 and 7200 RPM drives but I doubt
it will be great. I'd go for the drive that offers me the most storage for
the best price.
John
| |
| Barry Watzman 2003-12-25, 9:24 pm |
| There will be a slight speed difference. You can measure it with
benchmarks, but in most cases you won't notice it. However, it does
depend somewhat on what you are doing. Same for data transfer rate, but
RPM is probably more noticeable than data transfer rate.
PLAIN JOE wrote:
> "JBS" <noone@thisdontwork.net> wrote in message
> news:01GGb.8883$IM3.8466@newsread3.news.atl.earthlink.net...
>
>
> interface.
>
> What will be difference in a newer system, PC 133, running on a 5400 RPM
> disk or 7200 RPM?
>
>
| |
| Richard Ballard 2003-12-26, 12:23 am |
| In article <3FEB94AF.60007@neo.rr.com>,
Barry Watzman <WatzmanNOSPAM@neo.rr.com> writes:
>The speed of a drive is transparent to a computer, as is the data rate
>(e.g. ATA-66/100/133). Everything works with everything. Obviously,
>faster is faster, but as far as working, there are just no issues.
The faster the disk platter spins, the shorter the time until
the sector containing the beginning of a data file reaches
the disk heads. If a file is fragmented this effect is
magnified by the number of fragments.
To my knowledge, (ATA-E)IDE disks have less robust seek
algorithms than SCSI disks. To my knowledge, (ATA-E)IDE
full track buffers cannot solve the fragmentation delay problem.
Richard Ballard MSEE CNA4 KD0AZ
--
Consultant specializing in computer networks, imaging & security
Listed as rjballard in "Friends & Favorites" at www.amazon.com
Last book reviews: "Necronomicon" & "Necronomicon Spellbook"
by Ed Simon
| |
| Barry Watzman 2003-12-26, 3:24 am |
| There is no inherint difference between SCSI and IDE drives in terms of
seek algorithms or performance, however you can make a drive faster by
adding cost, and SCSI drives tend to be expensive, high-end drives while
IDE drives tend to be low-cost mass-market drives (although there are
both SCSI and IDE drives that blurr those distinctions). But nothing in
the two standards directly impacts any aspect of performance other than
data transfer rate.
Clearly the faster the drive spins the lower the rotational latency,
but, again, you could make a 15,000 rpm IDE drive or a 5400 rpm SCSI drive.
In practice, for typical desktop computing, none of the differences in
modern hard drives are terribly significant. However, if you are
talking about a server, or applications that are highly disk I/O bound,
then the differences can become significant.
Richard Ballard wrote:
> In article <3FEB94AF.60007@neo.rr.com>,
> Barry Watzman <WatzmanNOSPAM@neo.rr.com> writes:
>
>
>
>
> The faster the disk platter spins, the shorter the time until
> the sector containing the beginning of a data file reaches
> the disk heads. If a file is fragmented this effect is
> magnified by the number of fragments.
>
> To my knowledge, (ATA-E)IDE disks have less robust seek
> algorithms than SCSI disks. To my knowledge, (ATA-E)IDE
> full track buffers cannot solve the fragmentation delay problem.
>
> Richard Ballard MSEE CNA4 KD0AZ
> --
> Consultant specializing in computer networks, imaging & security
> Listed as rjballard in "Friends & Favorites" at www.amazon.com
> Last book reviews: "Necronomicon" & "Necronomicon Spellbook"
> by Ed Simon
>
| |
| Richard Ballard 2003-12-26, 11:24 am |
| In article <3FEBE07C.6090008@neo.rr.com>,
Barry Watzman <WatzmanNOSPAM@neo.rr.com> writes:
>There is no inherint difference between SCSI and IDE drives in
>terms of seek algorithms or performance, however you can make a
>drive faster by adding cost, and SCSI drives tend to be expensive,
>high-end drives while IDE drives tend to be low-cost mass-market
>drives (although there are both SCSI and IDE drives that blurr
>those distinctions). But nothing in the two standards directly
>impacts any aspect of performance other than
>data transfer rate.
People have been looking at disk seek algorithms since the
bad old days of the IBM 3330 dismountable disk packs.
Disk controller complexity is one inherent difference between
IDE hard disk drives and SCSI hard disk drives. IDE disk controllers
are simpler, and to my knowledge perform data access requests on
a first-in-first-out (FIFO) basis. SCSI disk controllers typically
employ so-called "intelligent" data access algorithms designed to
minimize disk head thrashing and to minimize the disk head radial
positioning delay that occurs between data retrievals. The
resulting SCSI accesses are not necessarily FIFO, but SCSI
controllers sort the resulting data stream and realize higher data
retrieval bandwidth and less head positioning mechanism
wear, all other things being equal.
To illustrate the difference, consider the (catastrophic) case of
a large data file that due to fragmentation is dispersed among
sector clusters located at both the inner tracks and the outer
tracks of the disk platter. With bad luck in data sector location
and original recording sequence, the FIFO IDE disk controller
causes the IDE disk head to "thrash" (vibrate) between the outer
platter tracks and the inner platter tracks, accessing the data
in its (catastrophic) FIFO order. A SCSI controller, on the
other hand, would first access (out of requested sequence) all
of the sector clusters on the outer platter tracks, then would
access (out of requested sequence) all of the sector clusters
on the inner platter tracks. A SCSI disk controller's intelligent
seek algorithm would avoid having the disk heads "thrash" (vibrate)
between the disk platters' inner and outer tracks.
In non-catastrophic examples, the typical SCSI controller
"elevator seek algorithm" causes the SCSI hard disk heads
to sweep the disk platter surfaces like automobile windshield
wipers, a strategy that is quick and is fairly wear-free and
vibration-free for the SCSI disk head positioning mechanism.
SCSI disk controller seek algorithms are decades-old,
mature technology.
>Clearly the faster the drive spins the lower the rotational latency,
>but, again, you could make a 15,000 rpm IDE drive or a 5400 rpm
>SCSI drive.
The faster the rotation the greater the platter frictionl bearing
wear. And in some cases I can imagine platter gyroscopic inertia
causing additional concerns about disk platter bearing wear in
disk platters spinning at very high RPM in a vibrating environment.
Do military tanks contain hard disk drives?
I consider ATA drives to be high speed brute force devices.
I consider SCSI drives to be optimized for reliability and
for long life. SCSI drives have higher initial cost, but that
higher initial cost is justified in some applications.
>In practice, for typical desktop computing, none of the differences
>in modern hard drives are terribly significant. However, if you
>are talking about a server, or applications that are highly disk
>I/O bound, then the differences can become significant.
Disk defragmentation is a concern in modern desktop operating
systems -- in some cases the workstation must go offline to
perform disk defragmentation. CAD and graphics arts applications
are I/O-intensive with big file sizes. And where do you draw the
line between a power user's desktop workstation and a
_lightly loaded_ file sharing, print serving and application server?
[color=blue]
>Richard Ballard wrote:
>
Richard Ballard MSEE CNA4 KD0AZ
--
Consultant specializing in computer networks, imaging & security
Listed as rjballard in "Friends & Favorites" at www.amazon.com
Last book reviews: "Necronomicon" & "Necronomicon Spellbook"
by Ed Simon
|
|
|
|
|