How NOT to clean your hard disk platterhttp://forum.hddguru.com/how-clean-hard-drive-disk-platter-t17515.html#p134546
Concerning the fallacious method of 'cleaning' your hard drive.
Junior0409 wrote: Wow, they were actually serious.
They were serious, but the advice they offered is more than thirty years old and the equipment they are referring to can now only be found in on display in a museum.
The internet is an invaluable tool, but bad information is out there, so the reader must remain informed or suffer the consequences of believing anything they perceive to be accurate
The media, especially the print media but sometimes even on-line, can be guilty of publishing erroneous or misleading information.
Back in the '80s, I was a DIY user and did a lot of my own repairs. I did replace platters on disk drives, but those behemoths used 18-inch diameter platters with a capacity of 2.5mb.
Those fixed/removable drives also weighed well over 150 pounds and cost $18,000. Replacement platters cost only $100 or so if purchased on the open market.
If the heads crashed, I had an experienced technician do the repairs because they had the required equipment, training and experience.
In the '60s, the techs could and would fix bad spots on fixed-head drum drives by carefully scraping the oxidation off with a knife.
In those days of mainframes, I programmed IBM 360s and earlier 14xx series with core memory. The IBM 360 had up to 256k of semiconductor memory and the 1401/1440s contained a maximum of 16k core memory.
Core memory was a blast, it had to be strung in a latticework of doughnut-shaped iron ferrite cores. http://en.wikipedia.org/wiki/Core_memory
One doughnut represented a single bit, there were eight bits for data and the ninth for parity.
A cabinet the size of a dishwasher held 4k. However, with core memory, one could power off a 14xx for the night, then power it up the next day and the data in memory would still be there, ready to continue processing.
That was then, but since the days of the old 5MB MFM 5.25-inch full height hard disk drive, the drives are sealed
, for a reason.
The head-to-platter surface is so small, by comparison, a single particle of smoke appears as a boulder to a read-write head flying over a disk platter and one speck of dust looks like a tree.
Only an imbecile would open up a disk drive on his kitchen table. To open a delicate device outside the confines of a certified clean room, including the essential personal gear and equipment, immediately contaminates and renders a disk drive inoperable.
Nowadays one can replace a one-terabyte drive for a little more than fifty bucks, so the critical elements are the data
, which can only be protected by frequent backups, and the time
required to make that data current and available for service.
When a business ignores this critical element, that of the timeliness
of the data, they risk the loss of the business itself.
Customers will not pay a bill without an invoice from their supplier and if the supplier loses the data, or if it takes too long to reconstruct the data, the business usually fails from a lack of cash flow.
The average user rarely or never backs up their data, so when their disk drive fails, as it inevitably does, they go into panic mode. It is frustrating to know your data is there but still inaccessible.
More users are now using an internet service to keep their data backed up and that isn't a problem if it consists of a few hundred gigabytes, but even that can take quite a while to restore.
If one has several terabytes of data, they should consider redundant external drives, backed up frequently.
I recommend redundant drives because if one is backing up to the same drive continually, how do they know if it has inadvertently become corrupted but still seems to be usable?
If one is copying to their only backup disk and that disk has been corrupt for who knows how long, what happens when the primary drive fails and the user realizes their backup is corrupt also? What then?
First, that sinking feeling when one knows they may be in serious trouble. Then, upon discovery that their backup isn't what they thought they had for insurance, panic and despair sets in.
If they have any common sense they look for a solution, but time is not on their side, they have bills to pay and what of all the archival data that they may have lost?
On top of that, data recovery is very specialized and time consuming, therefore expensive, and usually when you can least afford it.
If one alternates between two or more external drives and uses the latest backed up drive as the current one, it should become obvious there is a problem and the previous backup will still be available, although the data may not be as current.
However, there is a golden rule to be followed here, that is when a problem with a backup occurs, the previous backup becomes possibly the ONLY backup you have, so it should be protected at all cost.
One should immediately obtain a replacement for the one with the problem and copy the now-precious previous backup to it and the new drive will then become the current drive. NEVER copy over your only good backup!!!
Using the 'alternating external drive' strategy of performing backups, the frequency of backups should also be considered.
Each person should decide for themselves how often a backup is made. Personally, I tend to back up before I perform a major operation, for instance, adding new software or making changes in drive configurations like partition changes or reorganizing folders.
Because I tend to do that frequently, for my purposes, backups are performed at least weekly or even daily.
I always maintain my operating system on a smaller drive which can be backed up, offline, in fifteen or twenty minutes. I tend to back up the entire drive as a rule, but also will back up only a partition if minor changes are necessary.
Partition backups take about five minutes or so and can usually be done without rebooting the system.
My data is a different situation since I have many terabytes of archival data. Because these backups can be done without rebooting the system, and because it is many times faster, I copy these to an image file on an internal drive.
I then unload the image file to one of my redundant external drives. This may take as much as 18 hours or so, but my system is up 24 hours per day and that takes few resources to perform.
If I ever have a need to interrupt the transfer, I can always simply restart it to the same drive as before. Remember the golden rule, NEVER to a previous backup which is considered as precious as gold.
Concerning the drive with the problem, one can tinker with it, send it off for recovery or throw it away.
Time is money and I have spent a lot of time just learning how data is structured on disk drives. If I were bold, I may swap out a drive's circuit board, purchased from a supplier who would provide the correct board and firmware for that particular drive. I would rely on his expertise and expect to pay for his service.
If I have kept my data sufficiently backed up, I can easily replace a large disk drive for a pittance, therefore there is no need to panic.
I can tinker with the failed disk when I decide to take the time and if it is a configuration problem, then I will have another spare for whenever I need it.
It would then be even better to have an additional disk for rotation in my backup set.