Skip to Content

Storage Grows in a Flash

The four-gigabyte Flash storage card in Apple’s iPod nano was obsolete before it hit store shelves.
November 30, 2005

If you cracked open an iPod nano, you’d find 16 flash chips inside, each capable of holding 250 megabytes of data, or room for about 1,000 songs. It’s pretty impressive. As soon as Apple rolled out its latest gadget, however, Samsung, which manufactures the chip, introduced a new flash chip that can hold eight times as much. So the iPod nano’s “cool” factor may have receded a bit.

But there’s a deeper trend here: the storage technology in consumer electronics devices may soon influence how computers themselves operate. Flash memory is coming to the PC.

Kevin Teixeira, a spokesperson for Intel, says data storage components, such as hard disks and flash chips, are actually outpacing Moore’s Law, the credo that predicts the number of transistors on a chip will double roughly every 18 months. At the same time, the demand for the iPod nano, smart phones, digital cameras, and other devices that use flash memory will keep driving down the price of flash memory components.

Unlike the spinning hard drive in today’s computers, as well as iPods from months ago, flash memory has no moving parts, making a smaller, more rugged gadget that’s also less prone to failure.

“Digital photography is possible because of flash,” explains Don Barnetson, associate director of flash marketing at Samsung. “Flash makes things smaller, take less power, and last longer.” Barnetson predicts that flash storage technology will bring features to cell phones that, two or three years ago, weren’t even available on your PC, such as storing and watching entire movies.

Flash can facilitate these operations because it stores data very differently than traditional hard drives. Right now, computers use a disk that stores individual bits, 1s and 0s, as a magnetic orientation in regions of the disk. The disk spins at 4,200 revolutions per minutes (in laptops), and the bits are read and written by a read-write head that resembles the arm of an old-fashioned LP player.

Because the disk spins, it can take up to 10 milliseconds to read a bit of data. Also, the motion drains energy from a battery and leaves the disk vulnerable to sudden movements that can damage a portion of the drive.

Flash avoids these problems because it is made of tiny transistors on a silicon chip. There are no moving parts, only moving electrons. Flash storage comes in two basic categories, NAND and NOR, where both names refer to the type of logic gates that allow the transport of bits. As Barnetson explains, data can be written faster to NAND than to NOR chips, and NAND chips also take up less space on a silicon wafer, and so are less expensive to produce.

Samsung specializes in NAND, and other companies are also seeing the benefits of this type of flash. On November 21, Intel and Micron Technology Inc. announced the formation of a spin-off company to manufacture NAND flash memory.

“In 2006, flash will start to penetrate the PC,” says Barnetson. Microsoft’s next-generation operating system, Vista, is designed to run best on PCs that include 128 megabytes of flash memory in addition to a traditional magnetic hard disk.

This “hybrid drive” will produce a number of noticeable benefits, says Barnetson. For one, information can be written to the flash drive, and then, about every 10 minutes, dumped from flash memory to the hard disk. This means the hard disk spins only periodically, reducing the amount of power consumed by 90 percent, according to Barnetson. And since a hard disk chews up about 10 percent of a computer’s power, this could translate into 30 more minutes on a laptop’s battery charge.

Moreover, because flash is “nonvolatile,” meaning it stores data even when the power is off, the flash memory in a computer can act as a backup, keeping files intact. Storage without power will also enable these hybrid machines to start up almost instantly. Barnetson notes that as a user is waiting for the hard drive to get up and running, the flash drive can be streaming information out of the cache (see “Starting Your Computer in a Flash”).

Barnetson believes that, while the cost of flash memory is currently too high to offer a 40-gigabyte laptop within the next few years, prices are decreasing by 35-40 percent per year. Devices such as tablet PCs that don’t need as much storage may be the first to arrive with all flash memory hard drives.

Furthermore, David Patterson, professor of electrical engineering at Stanford, suggests that laptops may not need all the storage space that users are currently accustomed to. If more data is stored in networks and accessed via the Internet, Patterson suggests, a 10-gigabyte laptop might suffice. In this case, the laptop’s memory could be reserved for programs that are rewritten less often, keeping the flash memory fresh.

But Patterson also doubts that Flash will entirely replace magnetic hard drives anytime soon. He notes that flash memory wears out after about 100,000 rewrites. This means it’s good for an iPod, where songs are updated every couple of days or so, but bad for software that has to write constantly to a computer’s hard drive.

Samsung has created a program that lessens the wear-and-tear on flash chips by evenly distributing data rewrites, so that one particular cell of the chip is not bombarded with information in rewriting, according to Barnetson. Even so, Patterson says, a cleverly written virus could wipe out an entire flash drive by taking advantage of this weakness.

And flash memory can’t do everything alone. Intel’s Teixeira notes that another issue is to increase bandwidth – the speed at which information is transferred from devices or streamed over networks. Join up flash memory with pumped-up bandwidth, Teixeira says, and “we’re heading toward a ubiquitous computing world where anything has intelligence.”

(For an annotated, warranty-voiding dissection of the iPod nano, see our “Hack” feature in the December 2005/January 2006 issue of Technology Review magazine.)

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

How scientists traced a mysterious covid case back to six toilets

When wastewater surveillance turns into a hunt for a single infected individual, the ethics get tricky.

The problem with plug-in hybrids? Their drivers.

Plug-in hybrids are often sold as a transition to EVs, but new data from Europe shows we’re still underestimating the emissions they produce.

It’s time to retire the term “user”

The proliferation of AI means we need a new word.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.