We’ve all done it before – purchased the cheapest priced HDD going, only to be devastated when it died, taking your data to the grave with it.
You can cross your fingers, pray, knock on wood, keep a rabbit’s foot on your keychain, or develop a ritual of other superstitious activities, but none of them will keep your hard drive from failing at some point in time. Although it’s a headache for you, the user, this inevitable reality keeps hard drive manufacturers and backup services in business. In order to avoid this failure from striking at a very inconvenient (and often costly) time, you must understand data reliability and how it’s measured.
One statistic that hard drive manufacturers use to measure their products’ reliabilities is called MTBF. For instance, Maxtor’s (pun intended) “theoretical MTBF” is computed based on what’s known about other drives with similar technical specifications to the one being released for sale. A “practical MTBF” can be computed within a year, based on information available from studies conducted during the year that examine the rate of crashed drives. It sounds complicated until you learn that the acronym simply stands for Mean Time Between Failures, or the average lifespan of a disk – from birth till death. Ranging anywhere from one to two million hours for the most recently developed HDDs, this simple unit helps manufacturers compare their statistical data to that of other companies. It’s important to understand how MTBF works, because if you do the math, 1,000,000 hours equates to 114 years, and realistically speaking, we can neither test a disk for that long, nor could we reasonably expect it to last this long. This is because the MTBF cannot be used to estimate a lifespan of a HDD; instead, the MTBF figure is used in combination with an analysis of a large amount of disks tested at once.
To calculate the MTBF, you sum up all the differences between downtime and uptime, and divide it by the overall number of failures. If you study this variable more closely, you will learn about MTBSA (Mean Time Between System Aborts), MTBUR (Mean Time Between Unit Replacements) and MTBCF (Mean Time Between Critical Failures). These numbers should also not be confused with MTTF (Mean Time To Failure), but that’s a separate topic.
Latest technological developments in smaller, more stable motors, resistors, wires, and other components have all made newer HDDs’ lifespan greater. Another factor that contributes to data reliability in hard drives is their capacity, and therefore its data density. Although it’s theoretically possible that larger-capacity hard drives fail more frequently than smaller hard drives, there are no statistical or experimental conclusions based on the occurrence of this issue. As with other electronics, a good way to measure a particular manufacturer’s reliability is to check out their warranty policies: the longer they’re willing to service your unit, the more confident the company is about its quality.
In today’s market, the traditional, 54-year-old hard drive is facing a new revival: it’s called Solid State Drives. First, they are smaller in size. Second, these have near-instant read speeds, because there are no moving parts, meaning Solid State Drives can be accessed instantly, whereas a conventional HDD needs to get up to speed to run correctly. Because of this, applications can be started faster. Third, they also are less likely to fail as quick as “old-school” HDDs, but come at a higher price, and don’t support as high of a capacity at this particular time. On the downside, SATA SSDs tend to write data on the SSD a bit slower. However, with doubled data density, faster start-up times, higher reliability, lack of noise during operation, and rigid enough construction to survive shocks like high vibrations and extreme temperatures, it’s definitely the next-generation hard drive – it will make sense to stop using spinning HDDs and switch to Solid State Drives (as soon as they get cheaper, when it would be cost-effective to store your OS and busy applications on the SSD, and keep the HDD for your data storage).
On average, you should replace your HDD every 5 to 6 years. It’s advised to get your HDDs changed to avoid failure that can ruin your important data permanently. No system cleaning software can prevent natural wear and tear that happen’s over time – and when it happens, there will be no one to blame but you; you’ve been warned.
