Hard Drive Failures

Are Drive Failures Increasing?

har

Introduction

Hard drive crashes are one of the most frustrating experiences one can have with a computer. The inability to read data off of the hard drive can render a computer useless. Even if the OS can run, the data may be inaccessible or damaged. The only way to recover from such a failure is to typically restore data from a backup onto a new drive with all the software installed from scratch. If no backup is available, then the data is either lost or will cost a lot for recovery services to retrieve. This article is going to take a look at what causes hard drive failures, if failures are becoming more frequent and what steps one can take to try and avoid problems in the event of a failure.

Hard Drive Basics

Before understanding what can cause a failure, it is important to know the basics of how a hard drive functions. A hard drive is essentially a large device with magnetic storage media that is encased on rigid platters. This allows the drive to store large amounts of data that can be accessed and written to very quickly. Every hard drive is comprised of several key components: case, drive motor, platters, drive heads and a logic board. The case provides protection for the drive in a sealed environment away from dust particles. The motor spins the drive up so the data can be read off of the platters. The platters hold the magnetic media that stores the actual data. The drive heads are used to read and write the data to the platters. Finally the logic board controls how the drive interfaces and talks to the rest of the computer system.

Common Drive Failures

The most common failure for a hard drive is something called a head crash. A head crash is any instance where the drive head manages to touch a platter. When this happens, the magnetic media will be etched off of the platter by the head and render both the data and the drive head inoperable. There is no clean recovery from such a failure. Another common failure comes from imperfections on the magnetic media. Any time that a sector on the disk fails to properly hold the magnetic alignment will cause data to be inaccessible. Typically drives will have a few of these located on the platter, but they are marked out of use by a low level format from the manufacturer. Later low level formats can be done to mark sectors as unusable so they will not be used, but this is a long process that erases all data from the drive. Mobile systems tended to be prone to platters that shattered. This was due to the fact that most hard drive platters are made of glass and were susceptible to shock. Most manufacturers have or are switching to other materials to prevent this from happening. If there is electrical problems with the logic board, data on the drive can become unreadable or damaged. This is due to the logic board being unable to properly communicate between the computer system and the hard drive.

MTBF

In order for consumers to get a good idea of the lifespan of a hard drive, a drive was rated by something called MTBF. This term stands for Mean Time Between Failure and is used to represent the length of time that 50 percent of drives would fail before and 50 percent would fail after. It is used to give an idea to a buyer as to the average amount of time the device will function for. This was typically listed by the manufacturers on all computer drives but in recent years it has been removed from all consumer drives. They are still listed for enterprise class hard drives.

Capacity vs. Reliability

Hard drive sizes have been increasing dramatically over the last few years. This is due to the increase in the density of data being stored on the platters and the number of platters that are being placed inside of a hard drive case. For example, most drives used to feature two or maybe three platters, but many now can have up to four total platters. This increase in the number of parts and the reduction in space has greatly reduced the tolerances that the drives have and increases the amount of possible chance of failure.

Are Drives More Prone to Failure Now?

A lot of this has to do with the construction and use of hard drives. Most consumer computers were used only few a few hours per day. This meant that the drives did not have as long of continuous use that increase factors such as heat and movement that can lead to failures. Computers are much more prevalent in our lives and are being used for longer periods. This means that drives are likely failing more frequently due to heavier use. After all, a computer used twice as long as another will generally have a hard drive fail twice as quickly. So this hasn’t really increased the failure rate. Of course, factors such as the increase in data density and number of platters may also be contributing the chances of hard drive failure. The more parts and the tighter the density of the data on the platters means that there are more things that can potentially go wrong to cause data loss or a failure. To counter this though, technology has been improving. Better motors, chemical composition of the media and other materials means that failures that used to happen due to these parts are less likely to occur. There is no hard evidence that failures are occurring more frequently. From my own personally experience, I have not seen an increase in the number of drives failing, but other people that I work with have seen a fair number of drives in their computers have problems. This is anecdotal evidence though. Warranties may be a good indicator of how the industry is dealing with reliability. After the dark days surrounding the infamous Deskstar problems, many manufacturers were reducing warranties. Before this the typical warranty was three years in length, but many companies switched to one year warranties. Now companies are typically offering three to five year long warranties meaning that they must have confidence in their drives as they are costly to replace.

What to Do in Case of Drive Failure?

The biggest problem with a drive failure is the amount of data that can be lost. With the increase in the number of digital devices that we use and resulting data being stored on our computer systems, it is much more disruptive to our lives to have it destroyed. Data recover from damaged drives can range from several hundred dollars to several thousand. Data recovery services aren’t flawless either. A head crash will likely remove the magnetic media from the platter destroying the data forever. There is no real way to prevent a drive failure either. Even the most reputable and reliable brand can have a drive that fails rapidly As a result, it is best to try and plan for an event that will cause the primary data drive to fail with data backups. There are a wide range of backup methods available to use. One simple tip I like to suggest to people is portable hard drives. They are fairly inexpensive and due to their limited use, are less likely to fail when properly stored and handled. External hard drives are available in the exact same capacities as the desktop drives because they often use the same drives. The key is to only use the drive when backing up data or restoring it. This reduces the amount of time it is used and lessen the chance of failure. Another option open to users is to build a desktop PC with a version of RAID that has data redundancy built in. The simplest form of RAID to setup is RAID 1 or mirroring. This requires a RAID controller and two identically sized hard drives. All data written to one drive is automatically mirrored to the other. In the event of a failure of one drive, the second drive will always have the data. Hard drives in general are very reliable, but with the increasing amount that we use our computers to store the data from our lives, the more catastrophic it becomes to have that data be lost from a drive failure. In general the drives are not failing any faster than they have in the past, but with the increasing use of the drives, they are reaching their expected lifespan sooner. Because of this, it is important to be prepared for such an incident by planning how to store your data in multiple locations to avoid data loss.

How To Troubleshoot a Noisy Hard Drive

Hard drives are usually nearly silent but some do make a muted clicking sound when they’re being accessed – this is normal. On the other hand, if you start hearing noises only occasionally or noises that you’ve never heard before – like clicking, grinding or squealing – your hard drive may be failing. The steps below will help you determine what to do next before all of your precious data is gone for good.

Here’s How:

1. Run free hard drive diagnostic software, already available on many PCs or available on the Internet. Additionally, more advanced diagnostic software is available for a cost from third party software developers.

Note: At best, diagnostic software will only mark the areas of the hard drive that are failing as “bad” and prevent the computer from using them in the future. It will not truly fix a hard drive that is physically failing.

2. If any corrections made by the diagnostics software do not temporarily resolve the hard drive noise, do a complete backup of your system and replace the hard drive immediately.

3. If the diagnostics software helps resolve the clicking, grinding or squealing noises keep in mind that this is only a temporary solution. Chances are, the hard drive will continue to fail until it is completely unusable. The permanent solution is to do a complete backup of your system and replace the hard drive as soon as possible.

Tips:

1. Since there is no good way to repair a failing hard drive, protecting your data by performing regular backups is essential. With an up-to-date backup, recovering from a hard drive failure is as simple as installing a new drive and restoring your data.

What is the churning sound I hear from my hard drive whenever it is retrieving data?

Let’s say you do something simple like double-click on the icon for a spreadsheet file. This simple act, on many computers, can take 20 or 30 seconds to complete, and all during that time the hard disk is churning away. The hard-disk access light flickers and the drive might make a whirring, whizzing or high-pitched whining noise. If the mechanism in the drive is loud, you definitely know that something is going on! You can see that there is an arm that holds the read-write heads. This arm can move the heads to tracks near the hub or near the edge of the disk. A normal hard disk is 5 inches (12.5 cm) or so in diameter. This arm, therefore, can move about 2 inches (5 cm) across the face of the disk. The speed at which this arm can move is astonishing. The arm is very light, and its actuator is powerful and precise. The arm can slide across the face of the disk hundreds of times per second if it needs to. If you think about how a speaker works, there is not much of a difference. A speaker is moving a lightweight cone back and forth hundreds of times per second to generate sound. As the hard- disk arm moves back and forth rapidly, it sets up vibrations that our ears hear as sounds. Why, when you click on a simple spreadsheet file, would the disk’s heads have to move so much (20 or 30 seconds worth of movement sometimes)? There are three things that cause all the movement:

· To start a spreadsheet application like Excel, the hard disk has to load the application itself along with a number of DLLs (dynamic link libraries) that support the application. The total size of all these different files might be 10 to 20 megabytes, and the files are scattered all over the disk. Loading 20 megabytes of data takes a lot of time and requires the disk head to move thousands of times to retrieve all the pieces.

· The data file itself has to load. The operating system (OS) has to move the head to the drive’s directory to find the folder, make sure the file name exists, and then discover the location of the file. Then the OS may need to read dozens of tracks scattered all over the drive to access the file.

· If the physical RAM is full, then during the loading process the OS will have to unload parts of physical RAM and save them to the paging file on the disk. So while the OS is trying to load the spreadsheet application and all the DLLs and the data file, it is at the same time trying to write millions of bytes of data to the paging file to make room for the new application. The drive head is moving all over the disk to accomplish these intermingled tasks.

· Altogether, clicking on a single icon may cause 40 or 50 megabytes of data to move between the drive and RAM, with the disk heads repositioning themselves thousands of times in the process. That is why you hear the drive “churning” — it’s doing a lot of work!

Introduction

Hard drive crashes are one of the most frustrating experiences one can have with a computer. The inability to read data off of the hard drive can render a computer useless. Even if the OS can run, the data may be inaccessible or damaged. The only way to recover from such a failure is to typically restore data from a backup onto a new drive with all the software installed from scratch. If no backup is available, then the data is either lost or will cost a lot for recovery services to retrieve. This article is going to take a look at what causes hard drive failures, if failures are becoming more frequent and what steps one can take to try and avoid problems in the event of a failure.

Hard Drive Basics

Before understanding what can cause a failure, it is important to know the basics of how a hard drive functions. A hard drive is essentially a large device with magnetic storage media that is encased on rigid platters. This allows the drive to store large amounts of data that can be accessed and written to very quickly. Every hard drive is comprised of several key components: case, drive motor, platters, drive heads and a logic board. The case provides protection for the drive in a sealed environment away from dust particles. The motor spins the drive up so the data can be read off of the platters. The platters hold the magnetic media that stores the actual data. The drive heads are used to read and write the data to the platters. Finally the logic board controls how the drive interfaces and talks to the rest of the computer system.

Common Drive Failures

The most common failure for a hard drive is something called a head crash. A head crash is any instance where the drive head manages to touch a platter. When this happens, the magnetic media will be etched off of the platter by the head and render both the data and the drive head inoperable. There is no clean recovery from such a failure. Another common failure comes from imperfections on the magnetic media. Any time that a sector on the disk fails to properly hold the magnetic alignment will cause data to be inaccessible. Typically drives will have a few of these located on the platter, but they are marked out of use by a low level format from the manufacturer. Later low level formats can be done to mark sectors as unusable so they will not be used, but this is a long process that erases all data from the drive. Mobile systems tended to be prone to platters that shattered. This was due to the fact that most hard drive platters are made of glass and were susceptible to shock. Most manufacturers have or are switching to other materials to prevent this from happening. If there is electrical problems with the logic board, data on the drive can become unreadable or damaged. This is due to the logic board being unable to properly communicate between the computer system and the hard drive.

MTBF

In order for consumers to get a good idea of the lifespan of a hard drive, a drive was rated by something called MTBF. This term stands for Mean Time Between Failure and is used to represent the length of time that 50 percent of drives would fail before and 50 percent would fail after. It is used to give an idea to a buyer as to the average amount of time the device will function for. This was typically listed by the manufacturers on all computer drives but in recent years it has been removed from all consumer drives. They are still listed for enterprise class hard drives.

Capacity vs. Reliability

Hard drive sizes have been increasing dramatically over the last few years. This is due to the increase in the density of data being stored on the platters and the number of platters that are being placed inside of a hard drive case. For example, most drives used to feature two or maybe three platters, but many now can have up to four total platters. This increase in the number of parts and the reduction in space has greatly reduced the tolerances that the drives have and increases the amount of possible chance of failure.

Are Drives More Prone to Failure Now?

A lot of this has to do with the construction and use of hard drives. Most consumer computers were used only few a few hours per day. This meant that the drives did not have as long of continuous use that increase factors such as heat and movement that can lead to failures. Computers are much more prevalent in our lives and are being used for longer periods. This means that drives are likely failing more frequently due to heavier use. After all, a computer used twice as long as another will generally have a hard drive fail twice as quickly. So this hasn’t really increased the failure rate. Of course, factors such as the increase in data density and number of platters may also be contributing the chances of hard drive failure. The more parts and the tighter the density of the data on the platters means that there are more things that can potentially go wrong to cause data loss or a failure. To counter this though, technology has been improving. Better motors, chemical composition of the media and other materials means that failures that used to happen due to these parts are less likely to occur. There is no hard evidence that failures are occurring more frequently. From my own personally experience, I have not seen an increase in the number of drives failing, but other people that I work with have seen a fair number of drives in their computers have problems. This is anecdotal evidence though. Warranties may be a good indicator of how the industry is dealing with reliability. After the dark days surrounding the infamous Deskstar problems, many manufacturers were reducing warranties. Before this the typical warranty was three years in length, but many companies switched to one year warranties. Now companies are typically offering three to five year long warranties meaning that they must have confidence in their drives as they are costly to replace.

What to Do in Case of Drive Failure?

The biggest problem with a drive failure is the amount of data that can be lost. With the increase in the number of digital devices that we use and resulting data being stored on our computer systems, it is much more disruptive to our lives to have it destroyed. Data recover from damaged drives can range from several hundred dollars to several thousand. Data recovery services aren’t flawless either. A head crash will likely remove the magnetic media from the platter destroying the data forever.

There is no real way to prevent a drive failure either. Even the most reputable and reliable brand can have a drive that fails rapidly As a result, it is best to try and plan for an event that will cause the primary data drive to fail with data backups. There are a wide range of backup methods available to use.

One simple tip I like to suggest to people is portable hard drives. They are fairly inexpensive and due to their limited use, are less likely to fail when properly stored and handled. External hard drives are available in the exact same capacities as the desktop drives because they often use the same drives. The key is to only use the drive when backing up data or restoring it. This reduces the amount of time it is used and lessen the chance of failure. Another option open to users is to build a desktop PC with a version of RAID that has data redundancy built in. The simplest form of RAID to setup is RAID 1 or mirroring. This requires a RAID controller and two identically sized hard drives. All data written to one drive is automatically mirrored to the other. In the event of a failure of one drive, the second drive will always have the data. Hard drives in general are very reliable, but with the increasing amount that we use our computers to store the data from our lives, the more catastrophic it becomes to have that data be lost from a drive failure. In general the drives are not failing any faster than they have in the past, but with the increasing use of the drives, they are reaching their expected lifespan sooner. Because of this, it is important to be prepared for such an incident by planning how to store your data in multiple locations to avoid data loss.

How To Troubleshoot a Noisy Hard Drive

Hard drives are usually nearly silent but some do make a muted clicking sound when they’re being accessed – this is normal. On the other hand, if you start hearing noises only occasionally or noises that you’ve never heard before – like clicking, grinding or squealing – your hard drive may be failing. The steps below will help you determine what to do next before all of your precious data is gone for good.

Here’s How:

1. Run free hard drive diagnostic software, already available on many PCs or available on the Internet. Additionally, more advanced diagnostic software is available for a cost from third party software developers.

Note: At best, diagnostic software will only mark the areas of the hard drive that are failing as “bad” and prevent the computer from using them in the future. It will not truly fix a hard drive that is physically failing.

2. If any corrections made by the diagnostics software do not temporarily resolve the hard drive noise, do a complete backup of your system and replace the hard drive immediately.

3. If the diagnostics software helps resolve the clicking, grinding or squealing noises keep in

mind that this is only a temporary solution. Chances are, the hard drive will continue to fail until it is completely unusable. The permanent solution is to do a complete backup of your system and replace the hard drive as soon as possible.

Tips:

1. Since there is no good way to repair a failing hard drive, protecting your data by performing regular backups is essential. With an up-to-date backup, recovering from a hard drive failure is as simple as installing a new drive and restoring your data.

What is the churning sound I hear from my hard drive whenever it is retrieving data?

Let’s say you do something simple like double-click on the icon for a spreadsheet file. This simple act, on many computers, can take 20 or 30 seconds to complete, and all during that time the hard disk is churning away. The hard-disk access light flickers and the drive might make a whirring, whizzing or high-pitched whining noise. If the mechanism in the drive is loud, you definitely know that something is going on! You can see that there is an arm that holds the read-write heads. This arm can move the heads to tracks near the hub or near the edge of the disk. A normal hard disk is 5 inches (12.5 cm) or so in diameter. This arm, therefore, can move about 2 inches (5 cm) across the face of the disk. The speed at which this arm can move is astonishing. The arm is very light, and its actuator is powerful and precise. The arm can slide across the face of the disk hundreds of times per second if it needs to. If you think about how a speaker works, there is not much of a difference. A speaker is moving a lightweight cone back and forth hundreds of times per second to generate sound. As the hard- disk arm moves back and forth rapidly, it sets up vibrations that our ears hear as sounds. Why, when you click on a simple spreadsheet file, would the disk’s heads have to move so much (20 or 30 seconds worth of movement sometimes)? There are three things that cause all the movement:

· To start a spreadsheet application like Excel, the hard disk has to load the application itself along with a number of DLLs (dynamic link libraries) that support the application. The total size of all these different files might be 10 to 20 megabytes, and the files are scattered all over the disk. Loading 20 megabytes of data takes a lot of time and requires the disk head to move thousands of times to retrieve all the pieces.

· The data file itself has to load. The operating system (OS) has to move the head to the drive’s directory to find the folder, make sure the file name exists, and then discover the location of the file. Then the OS may need to read dozens of tracks scattered all over the drive to access the file.

· If the physical RAM is full, then during the loading process the OS will have to unload parts of physical RAM and save them to the paging file on the disk. So while the OS is trying to load the spreadsheet application and all the DLLs and the data file, it is at the same time trying to write millions of bytes of data to the paging file to make room for the new application. The drive head is moving all over the disk to accomplish these intermingled tasks.

· Altogether, clicking on a single icon may cause 40 or 50 megabytes of data to move between the drive and RAM, with the disk heads repositioning themselves thousands of times in the process. That is why you hear the drive “churning” — it’s doing a lot of work!

You may also like...

Leave a Reply

Your email address will not be published. Required fields are marked *