For many of us still on the stage, our earliest exposure to “memory storage” had to do with the number of songs we could get on a “45” or “33” rpm record. I never bought a “45” because they only had one song per side. However, each “33” had about 5 songs per side (around 15 minutes), and I always thought that was a better deal. And so it was through the age of cassette tapes in the 1970s, CD’s in the 1980s, and on to wherever we are now.
My first brush with computer memory storage was the 80-column punch cards used in the 1970s. Hundreds, no, thousands of punch cards. We used to haul around boxes of those things with 2000 cards per box. Those I understood. And it turns out, the basic function of the punch card was the same as that of the hard disk in your computer: to encode either a “zero” or a “1” in various combinations. Each “1” or “0” was (and is) called a “bit”. A string of 1s and 0s eight bits long comprises a “byte”, which represents one character, such as a letter or a number. (I find it easier to think of a byte as “word” written in an alphabet of only 0s and 1s, but my physicist son has tactfully pointed out that that is just not correct.). A typical book of 500 pages is about one million bytes—a “megabyte”. So is a 4-megapixel JPEG image. Or one minute of 128 kbit/sec MP3 compressed music. You get the idea.
With increased information being stored, our vocabulary has increased from megabytes to “gigabytes,” each of which is 1000 megabytes (a billion bytes). The next multiple of a thousand is a “terabyte”, and so on. A “yottabyte” has 24 zeros. I don’t even want to know how many punch cards that would be. Fortunately, bytes now are encoded magnetically (credit card), optically (CD), or on a semiconductor (computer, iPod).
It is interesting to take a peak over the horizon towards new computer storage media. It seems that the likeliest future contenders will utilize quantum mechanics or DNA. Quantum computing allocates 0s and 1s to the spin direction of individual electrons, atoms, photons, or molecules. They spin one way and they are a 0. Another way they are a 1. And, for another new vocabulary word, if it is both a 0 and 1, it is a “qubit”. Don’t ask me how an electron can simultaneously have two different spins. I don’t think physicists know either—just one of the spooky things that have emerged from small particle physics. BUT, they have actually made elements of a quantum computer, and Australians claim that quantum computing will be here in 5-10 years.
Now for the good stuff--DNA computing! DNA is composed of 4 bases—A, T, G, and C. It is the basic genetic component for all of life—bacteria to humans. The sequence of the bases forms a genetic code. What researchers have done is to assign the number 0 to A or C, and 1 to T or G. So the sequence AAG is 001. So is CCG. The letter “F”, for example, would receive a unique string of eight bases, and comprise one byte of information. In order that the stored information can be retrieved, a string of bases in the nature of a “bar code” is added at the start of the sequence to indicate where in the text that specific letter shows up.
In a paper published in August, Harvard researchers report that they used this technique, plus the most modern methods of DNA sequencing, replication, and reading, to encode a 53,000-word book with 11 images (5.2 megabits) onto 55,000 DNA strands. They were subsequently able to “read” these back and voila—recreate the book.
You might well ask what is the point of going through all this effort to devise a method that uses DNA to accomplish something that could be done easily on your laptop? Mostly it has to do with storage capability: 70 billion DNA copies of our 500-page book would fit onto an area the size of your fingernail. The Harvard researchers estimate that every bit of human-created digital information currently in existence could be stored in only 4 grams of DNA, a feat made possible, in part, by the fact that this is three-dimensional storage. Think DNA in a beaker, as opposed to molecules on the flat discs of your hard drive. Another advantage is that DNA is quite stable—so stable in fact that researchers have been able to isolate intact DNA from extinct mammoths. And finally, DNA can be copied with high reliability, as is going on within your cells this very moment.
So how does this compare with quantum computing? The expectation, as I understand it, is that quantum computing will be faster, but DNA information storage will be hugely greater. And what about a quantum computer linked with a DNA storage system? Wow! I hope I live that long.
And how does DNA memory compare to other sorts of storage such as flash drives and CDs? DNA memory>quantum holography>bacteria>flash memory>DVDs, with DNA memory being 1000 times greater per unit of volume than quantum memory and 100 billion times greater than a CD.
It turns out that, of course, patent applications have been filed in the area of DNA computing. Not many, but there are a few. Since patents only last 20 years from date of filing, an application filed today will be worthless unless a commercial product can be developed in the next 20 years. One may ask how many sales of DNA storage media could be made—that is, could this become a consumer product? I doubt it. One sale to Google to store all information known to mankind? Maybe. I’m not sure I’d be willing to spend $50,000 or so to get a patent in this field.
But then again, in the early 1990’s, I couldn’t see the need for a 20 MB hard drive on an Apple SE either.
My first brush with computer memory storage was the 80-column punch cards used in the 1970s. Hundreds, no, thousands of punch cards. We used to haul around boxes of those things with 2000 cards per box. Those I understood. And it turns out, the basic function of the punch card was the same as that of the hard disk in your computer: to encode either a “zero” or a “1” in various combinations. Each “1” or “0” was (and is) called a “bit”. A string of 1s and 0s eight bits long comprises a “byte”, which represents one character, such as a letter or a number. (I find it easier to think of a byte as “word” written in an alphabet of only 0s and 1s, but my physicist son has tactfully pointed out that that is just not correct.). A typical book of 500 pages is about one million bytes—a “megabyte”. So is a 4-megapixel JPEG image. Or one minute of 128 kbit/sec MP3 compressed music. You get the idea.
With increased information being stored, our vocabulary has increased from megabytes to “gigabytes,” each of which is 1000 megabytes (a billion bytes). The next multiple of a thousand is a “terabyte”, and so on. A “yottabyte” has 24 zeros. I don’t even want to know how many punch cards that would be. Fortunately, bytes now are encoded magnetically (credit card), optically (CD), or on a semiconductor (computer, iPod).
It is interesting to take a peak over the horizon towards new computer storage media. It seems that the likeliest future contenders will utilize quantum mechanics or DNA. Quantum computing allocates 0s and 1s to the spin direction of individual electrons, atoms, photons, or molecules. They spin one way and they are a 0. Another way they are a 1. And, for another new vocabulary word, if it is both a 0 and 1, it is a “qubit”. Don’t ask me how an electron can simultaneously have two different spins. I don’t think physicists know either—just one of the spooky things that have emerged from small particle physics. BUT, they have actually made elements of a quantum computer, and Australians claim that quantum computing will be here in 5-10 years.
Now for the good stuff--DNA computing! DNA is composed of 4 bases—A, T, G, and C. It is the basic genetic component for all of life—bacteria to humans. The sequence of the bases forms a genetic code. What researchers have done is to assign the number 0 to A or C, and 1 to T or G. So the sequence AAG is 001. So is CCG. The letter “F”, for example, would receive a unique string of eight bases, and comprise one byte of information. In order that the stored information can be retrieved, a string of bases in the nature of a “bar code” is added at the start of the sequence to indicate where in the text that specific letter shows up.
In a paper published in August, Harvard researchers report that they used this technique, plus the most modern methods of DNA sequencing, replication, and reading, to encode a 53,000-word book with 11 images (5.2 megabits) onto 55,000 DNA strands. They were subsequently able to “read” these back and voila—recreate the book.
You might well ask what is the point of going through all this effort to devise a method that uses DNA to accomplish something that could be done easily on your laptop? Mostly it has to do with storage capability: 70 billion DNA copies of our 500-page book would fit onto an area the size of your fingernail. The Harvard researchers estimate that every bit of human-created digital information currently in existence could be stored in only 4 grams of DNA, a feat made possible, in part, by the fact that this is three-dimensional storage. Think DNA in a beaker, as opposed to molecules on the flat discs of your hard drive. Another advantage is that DNA is quite stable—so stable in fact that researchers have been able to isolate intact DNA from extinct mammoths. And finally, DNA can be copied with high reliability, as is going on within your cells this very moment.
So how does this compare with quantum computing? The expectation, as I understand it, is that quantum computing will be faster, but DNA information storage will be hugely greater. And what about a quantum computer linked with a DNA storage system? Wow! I hope I live that long.
And how does DNA memory compare to other sorts of storage such as flash drives and CDs? DNA memory>quantum holography>bacteria>flash memory>DVDs, with DNA memory being 1000 times greater per unit of volume than quantum memory and 100 billion times greater than a CD.
It turns out that, of course, patent applications have been filed in the area of DNA computing. Not many, but there are a few. Since patents only last 20 years from date of filing, an application filed today will be worthless unless a commercial product can be developed in the next 20 years. One may ask how many sales of DNA storage media could be made—that is, could this become a consumer product? I doubt it. One sale to Google to store all information known to mankind? Maybe. I’m not sure I’d be willing to spend $50,000 or so to get a patent in this field.
But then again, in the early 1990’s, I couldn’t see the need for a 20 MB hard drive on an Apple SE either.
No comments:
Post a Comment