Happy Birthday to the Hard Drive - The Effects of Storage Increases and Kryder's Law
(Page 3 of 4 )
Those early hard disks (from 1956 to the early 1980s) were much more suited to a data center or large office environment than a home. As noted, they were large, cumbersome, and drew a lot of power. They were also somewhat delicate, so they couldn't be used in an industrial environment.
The IBM PC/XT featured the first hard disk widely used in the PC world. It held 10 MB. In terms of basic design, it isn't that different from the hard disks in use today. But the various incremental improvements made to that design over the past 50 years have been nothing short of breathtaking. From the original areal density of 2,000 bits per square inch, we reached 100 billion bits per square inch by 2005. Albert S. Hoagland, who belonged to the small group of IBM engineers who developed the first magnetic disk drive for data storage, puts the matter in perspective: "Storage density has increased by a factor of 50 million in 50 years."
So is there a law that governs storage capacity increase, as Moore's Law does for the capacity of CPUs? Some have tried to popularize Kryder's Law, named for Mark Kryder, chief technical officer of Seagate (who seems to have never formally stated the law). Judging from the numbers, storage capacity seems to double about every 23 months. This is slower than Moore's Law, but fast enough to bring about a number of changes.
Take the iPod and other MP3 players, for example. They would not have been possible without hard drives. TiVo also would not have been possible. Whole business models would not have been possible; without the hard drive, there would be no search engines. Amazon and eBay could not exist without hard drives. There would also be no web-based email such as Gmail. If we did have email, it would be so different from what we have today as to almost be unrecognizable. Can you imagine storing all your email on tape or floppies?
As storage capacity has increased, the cost of storage has gone down. That's one of the factors that have made the current technological explosion possible (along with the increases in CPU capacity). It has also changed the way programmers think. Back when storage space was more costly, programmers had to make certain choices about the best way for a program to utilize time (measured in CPU cycles) and space. Today, they still face the time-space equation, but they can solve it differently, making choices that would have been considered a wild waste of space 20 years ago.
Remember the Year 2000 bug? So many programs stored only two numbers for the year that many feared a huge meltdown as software in use by banks and other major companies failed to handle the change. But why did these programs store only two numbers instead of four? Because storing four numbers would have been too costly in terms of storage, so programmers made a trade off, never imagining that anyone would still be using those old programs by the time it mattered.
Next: The Future of Hard Drive Storage >>
More Opinions Articles
More By Terri Wells
/ 6 
